2020.03.31.txt

==========New Papers==========
1, TITLE:       DHP: Differentiable Meta Pruning via HyperNetworks
http://arxiv.org/abs/2003.13683
AUTHORS:        Yawei Li ; Shuhang Gu ; Kai Zhang ; Luc Van Gool ; Radu Timofte
COMMENTS:       Code will be available at https://github.com/ofsoundof/dhp
HIGHLIGHT:      In this paper, we propose a differentiable pruning method via hypernetworks for automatic network pruning and layer-wise configuration optimization.

2, TITLE:       Supervised and Unsupervised Detections for Multiple Object Tracking in Traffic Scenes: A Comparative Study
http://arxiv.org/abs/2003.13644
AUTHORS:        Hui-Lee Ooi ; Guillaume-Alexandre Bilodeau ; Nicolas Saunier
COMMENTS:       Accepted for ICIAR 2020
HIGHLIGHT:      In this paper, we propose a multiple object tracker, called MF-Tracker, that integrates multiple classical features (spatial distances and colours) and modern features (detection labels and re-identification features) in its tracking framework.

3, TITLE:       Designing Network Design Spaces
http://arxiv.org/abs/2003.13678
AUTHORS:        Ilija Radosavovic ; Raj Prateek Kosaraju ; Ross Girshick ; Kaiming He ; Piotr Dollár
COMMENTS:       CVPR 2020
HIGHLIGHT:      In this work, we present a new network design paradigm.

4, TITLE:       Detection of 3D Bounding Boxes of Vehicles Using Perspective Transformation for Accurate Speed Measurement
http://arxiv.org/abs/2003.13137
AUTHORS:        Viktor Kocur ; Milan Ftáčnik
COMMENTS:       Submitted to Machine Vision and Applications
HIGHLIGHT:      Main contribution of this paper is an improved construction of the perspective transformation which is more robust and fully automatic and an extended experimental evaluation of speed estimation.

5, TITLE:       Learning a Weakly-Supervised Video Actor-Action Segmentation Model with a Wise Selection
http://arxiv.org/abs/2003.13141
AUTHORS:        Jie Chen ; Zhiheng Li ; Jiebo Luo ; Chenliang Xu
COMMENTS:       11 pages, 8 figures, cvpr-2020 supplementary video: https://youtu.be/CX1hEOV9tlo
HIGHLIGHT:      To overcome these challenges, we propose a general Weakly-Supervised framework with a Wise Selection of training samples and model evaluation criterion (WS^2).

6, TITLE:       Defect segmentation: Mapping tunnel lining internal defects with ground penetrating radar data using a convolutional neural network
http://arxiv.org/abs/2003.13120
AUTHORS:        Senlin Yang ; Zhengfang Wang ; Jing Wang ; Anthony G. Cohn ; Jiaqi Zhang ; Peng Jiang ; Peng Jiang ; Qingmei Sui
COMMENTS:       24 pages,11 figures
HIGHLIGHT:      This research proposes a Ground Penetrating Radar (GPR) data processing method for non-destructive detection of tunnel lining internal defects, called defect segmentation.

7, TITLE:       NMS by Representative Region: Towards Crowded Pedestrian Detection by Proposal Pairing
http://arxiv.org/abs/2003.12729
AUTHORS:        Xin Huang ; Zheng Ge ; Zequn Jie ; Osamu Yoshie
COMMENTS:       Accepted by CVPR2020. The first two authors contributed equally, and are listed in alphabetical order
HIGHLIGHT:      To avoid such a dilemma, this paper proposes a novel Representative Region NMS approach leveraging the less occluded visible parts, effectively removing the redundant boxes without bringing in many false positives.

8, TITLE:       Inferring Semantic Information with 3D Neural Scene Representations
http://arxiv.org/abs/2003.12673
AUTHORS:        Amit Kohli ; Vincent Sitzmann ; Gordon Wetzstein
COMMENTS:       Project page can be found in https://www.computationalimaging.org/publications/
HIGHLIGHT:      Motivated by this ability of biological vision, we demonstrate that 3D-structure-aware representation learning leads to multi-modal representations that enable 3D semantic segmentation with extremely limited, 2D-only supervision.

9, TITLE:       Exploit Clues from Views: Self-Supervised and Regularized Learning for Multiview Object Recognition
http://arxiv.org/abs/2003.12735
AUTHORS:        Chih-Hui Ho ; Bo Liu ; Tz-Ying Wu ; Nuno Vasconcelos
COMMENTS:       Accepted to CVPR2020
HIGHLIGHT:      In this work, the problem of multiview self-supervised learning (MV-SSL) is investigated, where only image to object association is given.

10, TITLE:      Using the Split Bregman Algorithm to Solve the Self-Repelling Snake Model
http://arxiv.org/abs/2003.12693
AUTHORS:        Huizhu Pan ; Jintao Song ; Wanquan Liu ; Ling Li ; Guanglu Zhou ; Lu Tan ; Shichu Chen
HIGHLIGHT:      In our paper,we propose an alternative solution to the SR using the Split Bregman method.Our algorithm breaks the problem down into simpler subproblems to use lower-order evolution equations and approximation schemes.

11, TITLE:      Semantically Mutil-modal Image Synthesis
http://arxiv.org/abs/2003.12697
AUTHORS:        Zhen Zhu ; Zhiliang Xu ; Ansheng You ; Xiang Bai
COMMENTS:       To appear in CVPR 2020
HIGHLIGHT:      In this paper, we focus on semantically multi-modal image synthesis (SMIS) task, namely, generating multi-modal images at the semantic level.

12, TITLE:      Best Practices for Implementing FAIR Vocabularies and Ontologies on the Web
http://arxiv.org/abs/2003.13084
AUTHORS:        Daniel Garijo ; María Poveda-Villalón
COMMENTS:       16 pages, 4 figures
HIGHLIGHT:      In this chapter we describe guidelines and best practices for creating accessible, understandable and reusable ontologies on the Web, using standard practices and pointing to existing tools and frameworks developed by the Semantic Web community.

13, TITLE:      OCmst: One-class Novelty Detection using Convolutional Neural Network and Minimum Spanning Trees
http://arxiv.org/abs/2003.13524
AUTHORS:        Riccardo La Grassa ; Ignazio Gallo ; Nicola Landro
COMMENTS:       16 pages
HIGHLIGHT:      We present a novel model called One Class Minimum Spanning Tree (OCmst) for novelty detection problem that uses a Convolutional Neural Network (CNN) as deep feature extractor and graph-based model based on Minimum Spanning Tree (MST).

14, TITLE:      Predicting the Popularity of Micro-videos with Multimodal Variational Encoder-Decoder Framework
http://arxiv.org/abs/2003.12724
AUTHORS:        Yaochen Zhu ; Jiayi Xie ; Zhenzhong Chen
HIGHLIGHT:      In this paper, we propose a multimodal variational encoder-decoder (MMVED) framework for micro-video popularity prediction tasks.

15, TITLE:      Gradient-based Data Augmentation for Semi-Supervised Learning
http://arxiv.org/abs/2003.12824
AUTHORS:        Hiroshi Kaizuka
COMMENTS:       2 figures, 6 tables
HIGHLIGHT:      We propose an SSL method named MixGDA by combining various mixup methods and GDA.

16, TITLE:      Clickbait Detection using Multiple Categorization Techniques
http://arxiv.org/abs/2003.12961
AUTHORS:        Abinash Pujahari ; Dilip Singh Sisodia
COMMENTS:       11 pages, 7 figures, 4 tables to be published in Journal of Information Science
HIGHLIGHT:      This paper proposes a hybrid categorization technique for separating clickbait and non-clickbait articles by integrating different features, sentence structure, and clustering.

17, TITLE:      Analysing the Extent of Misinformation in Cancer Related Tweets
http://arxiv.org/abs/2003.13657
AUTHORS:        Rakesh Bal ; Sayan Sinha ; Swastika Dutta ; Risabh Joshi ; Sayan Ghosh ; Ritam Dutt
COMMENTS:       Proceedings of the 14th INTERNATIONAL CONFERENCE ON WEB AND SOCIAL MEDIA (ICWSM-20)
HIGHLIGHT:      In this work, we aim to tackle the misinformation spread in such platforms. We collect and present a dataset regarding tweets which talk specifically about cancer and propose an attention-based deep learning model for automated detection of misinformation along with its spread.

18, TITLE:      Empirical Comparison of Graph Embeddings for Trust-Based Collaborative Filtering
http://arxiv.org/abs/2003.13345
AUTHORS:        Tomislav Duricic ; Hussain Hussain ; Emanuel Lacic ; Dominik Kowald ; Denis Helic ; Elisabeth Lex
COMMENTS:       10 pages, Accepted as a full paper on the 25th International Symposium on Methodologies for Intelligent Systems (ISMIS'20)
HIGHLIGHT:      In this work, we study the utility of graph embeddings to generate latent user representations for trust-based collaborative filtering.

19, TITLE:      A Streaming On-Device End-to-End Model Surpassing Server-Side Conventional Model Quality and Latency
http://arxiv.org/abs/2003.12710
AUTHORS:        Tara N. Sainath ; Yanzhang He ; Bo Li ; Arun Narayanan ; Ruoming Pang ; Antoine Bruguier ; Shuo-yiin Chang ; Wei Li ; Raziel Alvarez ; Zhifeng Chen ; Chung-Cheng Chiu ; David Garcia ; Alex Gruenstein ; Ke Hu ; Minho Jin ; Anjuli Kannan ; Qiao Liang ; Ian McGraw ; Cal Peyser ; Rohit Prabhavalkar ; Golan Pundak ; David Rybach ; Yuan Shangguan ; Yash Sheth ; Trevor Strohman ; Mirko Visontai ; Yonghui Wu ; Yu Zhang ; Ding Zhao
COMMENTS:       In Proceedings of IEEE ICASSP 2020
HIGHLIGHT:      In this paper, we develop a first-pass Recurrent Neural Network Transducer (RNN-T) model and a second-pass Listen, Attend, Spell (LAS) rescorer that surpasses a conventional model in both quality and latency.

20, TITLE:      Variational Transformers for Diverse Response Generation
http://arxiv.org/abs/2003.12738
AUTHORS:        Zhaojiang Lin ; Genta Indra Winata ; Peng Xu ; Zihan Liu ; Pascale Fung
COMMENTS:       open domain dialogue
HIGHLIGHT:      Therefore, we propose the Variational Transformer (VT), a variational self-attentive feed-forward sequence model.

21, TITLE:      Unsupervised feature learning for speech using correspondence and Siamese networks
http://arxiv.org/abs/2003.12799
AUTHORS:        Petri-Johan Last ; Herman A. Engelbrecht ; Herman Kamper
COMMENTS:       5 pages, 3 figures, 2 tables; accepted to the IEEE Signal Processing Letters, (c) 2020 IEEE
HIGHLIGHT:      Here we compare two recent methods for frame-level acoustic feature learning.

22, TITLE:      HIN: Hierarchical Inference Network for Document-Level Relation Extraction
http://arxiv.org/abs/2003.12754
AUTHORS:        Hengzhu Tang ; Yanan Cao ; Zhenyu Zhang ; Jiangxia Cao ; Fang Fang ; Shi Wang ; Pengfei Yin
COMMENTS:       Accepted by PAKDD 2020
HIGHLIGHT:      In this paper, we propose a Hierarchical Inference Network (HIN) to make full use of the abundant information from entity level, sentence level and document level.

23, TITLE:      Serialized Output Training for End-to-End Overlapped Speech Recognition
http://arxiv.org/abs/2003.12687
AUTHORS:        Naoyuki Kanda ; Yashesh Gaur ; Xiaofei Wang ; Zhong Meng ; Takuya Yoshioka
COMMENTS:       Submitted to INTERSPEECH 2020
HIGHLIGHT:      This paper proposes serialized output training (SOT), a novel framework for multi-speaker overlapped speech recognition based on an attention-based encoder-decoder approach.

24, TITLE:      Orchestrating NLP Services for the Legal Domain
http://arxiv.org/abs/2003.12900
AUTHORS:        Julián Moreno-Schneider ; Georg Rehm ; Elena Montiel-Ponsoda ; Víctor Rodriguez-Doncel ; Artem Revenko ; Sotirios Karampatakis ; Maria Khvalchik ; Christian Sageder ; Jorge Gracia ; Filippo Maganza
COMMENTS:       Proceedings of the 12th Language Resources and Evaluation Conference (LREC 2020). To appear
HIGHLIGHT:      The key contribution of this paper is a workflow manager that enables the flexible orchestration of workflows based on a portfolio of Natural Language Processing and Content Curation services as well as a Multilingual Legal Knowledge Graph that contains semantic information and meaningful references to legal documents.

25, TITLE:      User Generated Data: Achilles' heel of BERT
http://arxiv.org/abs/2003.12932
AUTHORS:        Ankit Kumar ; Piyush Makhija ; Anuj Gupta
COMMENTS:       7 pages, 2 figures, 6 plots
HIGHLIGHT:      In this work we systematically show that when the data is noisy, there is a significant degradation in the performance of BERT.

26, TITLE:      Density-Aware Graph for Deep Semi-Supervised Visual Recognition
http://arxiv.org/abs/2003.13194
AUTHORS:        Suichan Li ; Bin Liu ; Dongdong Chen ; Qi Chu ; Lu Yuan ; Nenghai Yu
COMMENTS:       Accepted by CVPR2020
HIGHLIGHT:      Motivated by these limitations, this paper proposes to solve the SSL problem by building a novel density-aware graph, based on which the neighborhood information can be easily leveraged and the feature learning and label propagation can also be trained in an end-to-end way.

27, TITLE:      Incremental Learning In Online Scenario
http://arxiv.org/abs/2003.13191
AUTHORS:        Jiangpeng He ; Runyu Mao ; Zeman Shao ; Fengqing Zhu
HIGHLIGHT:      In this paper, we propose an incremental learning framework that can work in the challenging online learning scenario and handle both new classes data and new observations of old classes.

28, TITLE:      Adversarial Feature Hallucination Networks for Few-Shot Learning
http://arxiv.org/abs/2003.13193
AUTHORS:        Kai Li ; Yulun Zhang ; Kunpeng Li ; Yun Fu
COMMENTS:       To appear in CVPR 2020
HIGHLIGHT:      In this paper, we propose Adversarial Feature Hallucination Networks (AFHN) which is based on conditional Wasserstein Generative Adversarial networks (cWGAN) and hallucinates diverse and discriminative features conditioned on the few labeled samples.

29, TITLE:      Learning Interactions and Relationships between Movie Characters
http://arxiv.org/abs/2003.13158
AUTHORS:        Anna Kukleva ; Makarand Tapaswi ; Ivan Laptev
COMMENTS:       CVPR 2020 (Oral)
HIGHLIGHT:      In this work, we propose neural models to learn and jointly predict interactions, relationships, and the pair of characters that are involved.

30, TITLE:      Gradually Vanishing Bridge for Adversarial Domain Adaptation
http://arxiv.org/abs/2003.13183
AUTHORS:        Shuhao Cui ; Shuhui Wang ; Junbao Zhuo ; Chi Su ; Qingming Huang ; Qi Tian
COMMENTS:       CVPR2020
HIGHLIGHT:      In this paper, we equip adversarial domain adaptation with Gradually Vanishing Bridge (GVB) mechanism on both generator and discriminator.

31, TITLE:      Space-Time-Aware Multi-Resolution Video Enhancement
http://arxiv.org/abs/2003.13170
AUTHORS:        Muhammad Haris ; Greg Shakhnarovich ; Norimichi Ukita
COMMENTS:       To appear in CVPR2020
HIGHLIGHT:      We consider the problem of space-time super-resolution (ST-SR): increasing spatial resolution of video frames and simultaneously interpolating frames to increase the frame rate.

32, TITLE:      GPS-Net: Graph Property Sensing Network for Scene Graph Generation
http://arxiv.org/abs/2003.12962
AUTHORS:        Xin Lin ; Changxing Ding ; Jinquan Zeng ; Dacheng Tao
COMMENTS:       Accepted by CVPR 2020 as Oral. Code is available
HIGHLIGHT:      Accordingly, in this paper, we propose a Graph Property Sensing Network (GPS-Net) that fully explores these three properties for SGG.

33, TITLE:      Realistic Face Reenactment via Self-Supervised Disentangling of Identity and Pose
http://arxiv.org/abs/2003.12957
AUTHORS:        Xianfang Zeng ; Yusu Pan ; Mengmeng Wang ; Jiangning Zhang ; Yong Liu
HIGHLIGHT:      To alleviate the demand for manual annotations, in this paper, we propose a novel self-supervised hybrid model (DAE-GAN) that learns how to reenact face naturally given large amounts of unlabeled videos.

34, TITLE:      Global-Local Bidirectional Reasoning for Unsupervised Representation Learning of 3D Point Clouds
http://arxiv.org/abs/2003.12971
AUTHORS:        Yongming Rao ; Jiwen Lu ; Jie Zhou
COMMENTS:       Accepted to CVPR2020
HIGHLIGHT:      Based on this hypothesis, we propose to learn point cloud representation by bidirectional reasoning between the local structures at different abstraction hierarchies and the global shape without human supervision.

35, TITLE:      Spatial Attention Pyramid Network for Unsupervised Domain Adaptation
http://arxiv.org/abs/2003.12979
AUTHORS:        Congcong Li ; Dawei Du ; Libo Zhang ; Longyin Wen ; Tiejian Luo ; Yanjun Wu ; Pengfei Zhu
HIGHLIGHT:      To that end, in this paper, we design a new spatial attention pyramid network for unsupervised domain adaptation.

36, TITLE:      ClusterVO: Clustering Moving Instances and Estimating Visual Odometry for Self and Surroundings
http://arxiv.org/abs/2003.12980
AUTHORS:        Jiahui Huang ; Sheng Yang ; Tai-Jiang Mu ; Shi-Min Hu
COMMENTS:       2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
HIGHLIGHT:      We present ClusterVO, a stereo Visual Odometry which simultaneously clusters and estimates the motion of both ego and surrounding rigid clusters/objects.

37, TITLE:      DaST: Data-free Substitute Training for Adversarial Attacks
http://arxiv.org/abs/2003.12703
AUTHORS:        Mingyi Zhou ; Jing Wu ; Yipeng Liu ; Shuaicheng Liu ; Ce Zhu
COMMENTS:       Accepted by CVPR2020
HIGHLIGHT:      In this paper, we propose a data-free substitute training method (DaST) to obtain substitute models for adversarial black-box attacks without the requirement of any real data.

38, TITLE:      Deep reinforcement learning for large-scale epidemic control
http://arxiv.org/abs/2003.13676
AUTHORS:        Pieter Libin ; Arno Moonens ; Timothy Verstraeten ; Fabian Perez-Sanjines ; Niel Hens ; Philippe Lemey ; Ann Nowé
HIGHLIGHT:      For this reason, we investigate a deep reinforcement learning approach to automatically learn prevention strategies in the context of pandemic influenza.

39, TITLE:      GAN-based Priors for Quantifying Uncertainty
http://arxiv.org/abs/2003.12597
AUTHORS:        Dhruv V. Patel ; Assad A. Oberai
HIGHLIGHT:      In this work we demonstrate how the approximate distribution learned by a deep generative adversarial network (GAN) may be used as a prior in a Bayesian update to address both these challenges.

40, TITLE:      A Polynomial Degree Bound on Defining Equations of Non-rigid Matrices and Small Linear Circuits
http://arxiv.org/abs/2003.12938
AUTHORS:        Mrinal Kumar ; Ben Lee Volk
HIGHLIGHT:      We show that there is a defining equation of degree at most $\mathsf{poly}(n)$ for the (Zariski closure of the) set of the non-rigid matrices: that is, we show that for every large enough field $\mathbb{F}$, there is a non-zero $n^2$-variate polynomial $P \in \mathbb{F}(x_{1, 1}, \ldots, x_{n, n})$ of degree at most $\mathsf{poly}(n)$ such that every matrix $M$ which can be written as a sum of a matrix of rank at most $n/100$ and sparsity at most $n^2/100$ satisfies $P(M) = 0$.

41, TITLE:      A combinatorial MA-complete problem
http://arxiv.org/abs/2003.13065
AUTHORS:        Dorit Aharonov ; Alex B. Grilo
HIGHLIGHT:      In this note we define a natural combinatorial problem called SetCSP and prove its MA-completeness.

42, TITLE:      A faster algorithm for the FSSP in one-dimensional CA with multiple speeds
http://arxiv.org/abs/2003.13558
AUTHORS:        Thomas Worsch
COMMENTS:       11 pages, 3 figures
HIGHLIGHT:      In the present paper we derive lower bounds on possible synchronization times and describe an algorithm which is never slower and in some cases faster than the one by Manzoni and Umeo and which is close to a lower bound (up to a constant summand) in more cases.

43, TITLE:      Towards Supervised and Unsupervised Neural Machine Translation Baselines for Nigerian Pidgin
http://arxiv.org/abs/2003.12660
AUTHORS:        Orevaoghene Ahia ; Kelechi Ogueji
COMMENTS:       Accepted for the AfricaNLP Workshop, ICLR 2020
HIGHLIGHT:      This work aims to establish supervised and unsupervised neural machine translation (NMT) baselines between English and Nigerian Pidgin.

44, TITLE:      Domain-aware Visual Bias Eliminating for Generalized Zero-Shot Learning
http://arxiv.org/abs/2003.13261
AUTHORS:        Shaobo Min ; Hantao Yao ; Hongtao Xie ; Chaoqun Wang ; Zheng-Jun Zha ; Yongdong Zhang
COMMENTS:       Accepted by CVPR2020
HIGHLIGHT:      In this paper, we propose a novel Domain-aware Visual Bias Eliminating (DVBE) network that constructs two complementary visual representations, i.e., semantic-free and semantic-aligned, to treat seen and unseen domains separately.

45, TITLE:      Memory Aggregation Networks for Efficient Interactive Video Object Segmentation
http://arxiv.org/abs/2003.13246
AUTHORS:        Jiaxu Miao ; Yunchao Wei ; Yi Yang
COMMENTS:       Accepted to CVPR 2020. 10 pages, 9 figures
HIGHLIGHT:      In this work, we propose a unified framework, named Memory Aggregation Networks (MA-Net), to address the challenging iVOS in a more efficient way.

46, TITLE:      TapLab: A Fast Framework for Semantic Video Segmentation Tapping into Compressed-Domain Knowledge
http://arxiv.org/abs/2003.13260
AUTHORS:        Junyi Feng ; Songyuan Li ; Xi Li ; Fei Wu ; Qi Tian ; Ming-Hsuan Yang ; Haibin Ling
HIGHLIGHT:      In this paper, we rethink this problem from a different viewpoint: using knowledge contained in compressed videos.

47, TITLE:      Cross-domain Detection via Graph-induced Prototype Alignment
http://arxiv.org/abs/2003.12849
AUTHORS:        Minghao Xu ; Hang Wang ; Bingbing Ni ; Qi Tian ; Wenjun Zhang
COMMENTS:       Accepted as ORAL presentation at IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020
HIGHLIGHT:      To mitigate these problems, we propose a Graph-induced Prototype Alignment (GPA) framework to seek for category-level domain alignment via elaborate prototype representations.

48, TITLE:      Adversarial Robustness: From Self-Supervised Pre-Training to Fine-Tuning
http://arxiv.org/abs/2003.12862
AUTHORS:        Tianlong Chen ; Sijia Liu ; Shiyu Chang ; Yu Cheng ; Lisa Amini ; Zhangyang Wang
COMMENTS:       CVPR 2020
HIGHLIGHT:      We introduce adversarial training into self-supervision, to provide general-purpose robust pre-trained models for the first time.

49, TITLE:      One-Shot Domain Adaptation For Face Generation
http://arxiv.org/abs/2003.12869
AUTHORS:        Chao Yang ; Ser-Nam Lim
COMMENTS:       Accepted to CVPR 2020
HIGHLIGHT:      In this paper, we propose a framework capable of generating face images that fall into the same distribution as that of a given one-shot example.

50, TITLE:      Refined Plane Segmentation for Cuboid-Shaped Objects by Leveraging Edge Detection
http://arxiv.org/abs/2003.12870
AUTHORS:        Alexander Naumann ; Laura Dörr ; Niels Ole Salscheider ; Kai Furmans
HIGHLIGHT:      We propose a post-processing algorithm to align the segmented plane masks with edges detected in the image.

51, TITLE:      Adversarial Imitation Attack
http://arxiv.org/abs/2003.12760
AUTHORS:        Mingyi Zhou ; Jing Wu ; Yipeng Liu ; Shuaicheng Liu ; Xiang Zhang ; Ce Zhu
COMMENTS:       9 pages
HIGHLIGHT:      In this study, we propose a novel adversarial imitation attack.

52, TITLE:      NPENAS: Neural Predictor Guided Evolution for Neural Architecture Search
http://arxiv.org/abs/2003.12857
AUTHORS:        Chen Wei ; Chuang Niu ; Yiping Tang ; Jimin Liang
HIGHLIGHT:      In this paper, we propose two predictor based algorithms NPUBO and NPENAS for neural architecture search.

53, TITLE:      Unsupervised Deep Learning for MR Angiography with Flexible Temporal Resolution
http://arxiv.org/abs/2003.13096
AUTHORS:        Eunju Cha ; Hyungjin Chung ; Eung Yeop Kim ; Jong Chul Ye
HIGHLIGHT:      To address this problem, here we propose a novel unsupervised deep learning using optimal transport driven cycle-consistent generative adversarial network (cycleGAN).

54, TITLE:      Increasing negotiation performance at the edge of the network
http://arxiv.org/abs/2003.13668
AUTHORS:        Sam Vente ; Angelika Kimmig ; Alun Preece ; Federico Cerutti
COMMENTS:       Accepted for presentation at The 7th International Conference on Agreement Technologies (AT 2020)
HIGHLIGHT:      To improve this bottleneck, we introduce an extension to AOP called Alternating Constrained Offers Protocol (ACOP), in which agents can also express constraints to each other.

55, TITLE:      Cross-Domain Document Object Detection: Benchmark Suite and Method
http://arxiv.org/abs/2003.13197
AUTHORS:        Kai Li ; Curtis Wigington ; Chris Tensmeyer ; Handong Zhao ; Nikolaos Barmpalios ; Vlad I. Morariu ; Varun Manjunatha ; Tong Sun ; Yun Fu
COMMENTS:       To appear in CVPR 2020
HIGHLIGHT:      We investigate cross-domain DOD, where the goal is to learn a detector for the target domain using labeled data from the source domain and only unlabeled data from the target domain.

56, TITLE:      Learning Memory-guided Normality for Anomaly Detection
http://arxiv.org/abs/2003.13228
AUTHORS:        Hyunjong Park ; Jongyoun Noh ; Bumsub Ham
COMMENTS:       Accepted to CVPR 2020
HIGHLIGHT:      To address this problem, we present an unsupervised learning approach to anomaly detection that considers the diversity of normal patterns explicitly, while lessening the representation capacity of CNNs.

57, TITLE:      Learning to Learn Single Domain Generalization
http://arxiv.org/abs/2003.13216
AUTHORS:        Fengchun Qiao ; Long Zhao ; Xi Peng
COMMENTS:       In CVPR 2020 (13 pages including supplementary material). The source code and pre-trained models are publicly available at: https://github.com/joffery/M-ADA
HIGHLIGHT:      We propose a new method named adversarial domain augmentation to solve this Out-of-Distribution (OOD) generalization problem.

58, TITLE:      MetaFuse: A Pre-trained Fusion Model for Human Pose Estimation
http://arxiv.org/abs/2003.13239
AUTHORS:        Rongchang Xie ; Chunyu Wang ; Yizhou Wang
COMMENTS:       Accepted to CVPR2020
HIGHLIGHT:      In this work, we introduce MetaFuse, a pre-trained fusion model learned from a large number of cameras in the Panoptic dataset.

59, TITLE:      Deep Fashion3D: A Dataset and Benchmark for 3D Garment Reconstruction from Single Images
http://arxiv.org/abs/2003.12753
AUTHORS:        Heming Zhu ; Yu Cao ; Hang Jin ; Weikai Chen ; Dong Du ; Zhangye Wang ; Shuguang Cui ; Xiaoguang Han
COMMENTS:       23 pages, 9 figures. For project page, see https://kv2000.github.io/2020/03/25/deepFashion3DRevisited/
HIGHLIGHT:      To demonstrate the advantage of Deep Fashion3D, we propose a novel baseline approach for single-view garment reconstruction, which leverages the merits of both mesh and implicit representations.

60, TITLE:      Learning Invariant Representation for Unsupervised Image Restoration
http://arxiv.org/abs/2003.12769
AUTHORS:        Wenchao Du ; Hu Chen ; Hongyu Yang
HIGHLIGHT:      Instead, we propose an unsupervised learning method that explicitly learns invariant presentation from noisy data and reconstructs clear observations.

61, TITLE:      Actor-Transformers for Group Activity Recognition
http://arxiv.org/abs/2003.12737
AUTHORS:        Kirill Gavrilyuk ; Ryan Sanford ; Mehrsan Javan ; Cees G. M. Snoek
COMMENTS:       CVPR 2020
HIGHLIGHT:      We empirically study different ways to combine these representations and show their complementary benefits.

62, TITLE:      Trajectory Poisson multi-Bernoulli filters
http://arxiv.org/abs/2003.12767
AUTHORS:        Ángel F. García-Fernández ; Lennart Svensson ; Jason L. Williams ; Yuxuan Xia ; Karl Granström
HIGHLIGHT:      This paper presents two trajectory Poisson multi-Bernoulli (TPMB) filters for multi-target tracking: one to estimate the set of alive trajectories at each time step and another to estimate the set of all trajectories, which includes alive and dead trajectories, at each time step.

63, TITLE:      BiLingUNet: Image Segmentation by Modulating Top-Down and Bottom-Up Visual Processing with Referring Expressions
http://arxiv.org/abs/2003.12739
AUTHORS:        Ozan Arkan Can ; İlker Kesen ; Deniz Yuret
COMMENTS:       18 pages, 3 figures, submitted to ECCV 2020
HIGHLIGHT:      We present BiLingUNet, a state-of-the-art model for image segmentation using referring expressions.

64, TITLE:      PointGMM: a Neural GMM Network for Point Clouds
http://arxiv.org/abs/2003.13326
AUTHORS:        Amir Hertz ; Rana Hanocka ; Raja Giryes ; Daniel Cohen-Or
COMMENTS:       CVPR 2020 -- final version
HIGHLIGHT:      We present PointGMM, a neural network that learns to generate hGMMs which are characteristic of the shape class, and also coincide with the input point cloud.

65, TITLE:      Adaptive Reward-Poisoning Attacks against Reinforcement Learning
http://arxiv.org/abs/2003.12613
AUTHORS:        Xuezhou Zhang ; Yuzhe Ma ; Adish Singla ; Xiaojin Zhu
HIGHLIGHT:      We categorize such attacks by the infinity-norm constraint on $\delta_t$: We provide a lower threshold below which reward-poisoning attack is infeasible and RL is certified to be safe; we provide a corresponding upper threshold above which the attack is feasible.

66, TITLE:      MCFlow: Monte Carlo Flow Models for Data Imputation
http://arxiv.org/abs/2003.12628
AUTHORS:        Trevor W. Richardson ; Wencheng Wu ; Lei Lin ; Beilei Xu ; Edgar A. Bernal
HIGHLIGHT:      To that end, we propose MCFlow, a deep framework for imputation that leverages normalizing flow generative models and Monte Carlo sampling.

67, TITLE:      Learning medical triage from clinicians using Deep Q-Learning
http://arxiv.org/abs/2003.12828
AUTHORS:        Albert Buchard ; Baptiste Bouvier ; Giulia Prando ; Rory Beard ; Michail Livieratos ; Dan Busbridge ; Daniel Thompson ; Jonathan Richens ; Yuanzhao Zhang ; Adam Baker ; Yura Perov ; Kostis Gourgoulias ; Saurabh Johri
COMMENTS:       17 pages, 4 figures, 3 tables, preprint, in press
HIGHLIGHT:      In this work, we present a Deep Reinforcement Learning approach (a variant of DeepQ-Learning) to triage patients using curated clinical vignettes.

68, TITLE:      Coronavirus Optimization Algorithm: A bioinspired metaheuristic based on the COVID-19 propagation model
http://arxiv.org/abs/2003.13633
AUTHORS:        F. Martínez-Álvarez ; G. Asencio-Cortés ; J. F. Torres ; D. Gutiérrez-Avilés ; L. Melgar-García ; R. Pérez-Chacón ; C. Rubio-Escudero ; J. C. Riquelme ; A. Troncoso
COMMENTS:       26 pages, 3 figures
HIGHLIGHT:      A novel bioinspired metaheuristic is proposed in this work, simulating how the Coronavirus spreads and infects healthy people.

69, TITLE:      Re-purposing Heterogeneous Generative Ensembles with Evolutionary Computation
http://arxiv.org/abs/2003.13532
AUTHORS:        Jamal Toutouh ; Erik Hemberg ; Una-May O'Reily
COMMENTS:       Accepted as a full paper for the Genetic and Evolutionary Computation Conference - GECCO'20
HIGHLIGHT:      In this study, we apply two evolutionary algorithms (EAs) to create ensembles to re-purpose generative models, i.e., given a set of heterogeneous generators that were optimized for one objective (e.g., minimize Frechet Inception Distance), create ensembles of them for optimizing a different objective (e.g., maximize the diversity of the generated samples).

70, TITLE:      Extending Automated Deduction for Commonsense Reasoning
http://arxiv.org/abs/2003.13159
AUTHORS:        Tanel Tammet
COMMENTS:       19 pages, no figures
HIGHLIGHT:      Instead of devising new specialized logics we propose a framework of extensions to the mainstream resolution-based search methods to make these capable of performing search tasks for practical commonsense reasoning with reasonable efficiency.

71, TITLE:      Suphx: Mastering Mahjong with Deep Reinforcement Learning
http://arxiv.org/abs/2003.13590
AUTHORS:        Junjie Li ; Sotetsu Koyamada ; Qiwei Ye ; Guoqing Liu ; Chao Wang ; Ruihan Yang ; Li Zhao ; Tao Qin ; Tie-Yan Liu ; Hsiao-Wuen Hon
HIGHLIGHT:      We design an AI for Mahjong, named Suphx, based on deep reinforcement learning with some newly introduced techniques including global reward prediction, oracle guiding, and run-time policy adaptation.

72, TITLE:      Parallel Knowledge Transfer in Multi-Agent Reinforcement Learning
http://arxiv.org/abs/2003.13085
AUTHORS:        Yongyuan Liang ; Bangwei Li
HIGHLIGHT:      This paper proposes a novel knowledge transfer framework in MARL, PAT (Parallel Attentional Transfer).

73, TITLE:      Faster than FAST: GPU-Accelerated Frontend for High-Speed VIO
http://arxiv.org/abs/2003.13493
AUTHORS:        Balazs Nagy ; Philipp Foehn ; Davide Scaramuzza
COMMENTS:       Submitted to IEEE International Conference on Intelligent Robots and Systems (IROS), 2020. Open-source implementation available at https://github.com/uzh-rpg/vilib
HIGHLIGHT:      Our second contribution introduces an enhanced FAST feature detector that applies the aforementioned non-maxima suppression method.

74, TITLE:      SiTGRU: Single-Tunnelled Gated Recurrent Unit for Abnormality Detection
http://arxiv.org/abs/2003.13528
AUTHORS:        Habtamu Fanta ; Zhiwen Shao ; Lizhuang Ma
COMMENTS:       14 pages, 11 figures, 13 tables, this paper is accepted on Journal of Information Sciences
HIGHLIGHT:      In this paper, we propose a novel version of Gated Recurrent Unit (GRU), called Single Tunnelled GRU for abnormality detection.

75, TITLE:      A Comparison of Data Augmentation Techniques in Training Deep Neural Networks for Satellite Image Classification
http://arxiv.org/abs/2003.13502
AUTHORS:        Mohamed Abdelhack
HIGHLIGHT:      This study focuses on the topic of image augmentation in training of deep neural network classifiers.

76, TITLE:      Improved Gradient based Adversarial Attacks for Quantized Networks
http://arxiv.org/abs/2003.13511
AUTHORS:        Kartik Gupta ; Thalaiyasingam Ajanthan
HIGHLIGHT:      In this work, we systematically study the robustness of quantized networks against gradient based adversarial attacks and demonstrate that these quantized models suffer from gradient vanishing issues and show a fake sense of security.

77, TITLE:      Improving out-of-distribution generalization via multi-task self-supervised pretraining
http://arxiv.org/abs/2003.13525
AUTHORS:        Isabela Albuquerque ; Nikhil Naik ; Junnan Li ; Nitish Keskar ; Richard Socher
HIGHLIGHT:      We introduce a new self-supervised pretext task of predicting responses to Gabor filter banks and demonstrate that multi-task learning of compatible pretext tasks improves domain generalization performance as compared to training individual tasks alone.

78, TITLE:      LayoutMP3D: Layout Annotation of Matterport3D
http://arxiv.org/abs/2003.13516
AUTHORS:        Fu-En Wang ; Yu-Hsuan Yeh ; Min Sun ; Wei-Chen Chiu ; Yi-Hsuan Tsai
COMMENTS:       Annotation is available at https://github.com/fuenwang/LayoutMP3D
HIGHLIGHT:      To facilitate the learning algorithms for autonomous systems in indoor scenarios, we consider the Matterport3D dataset with their originally provided depth map ground truths and further release our annotations for layout ground truths from a subset of Matterport3D.

79, TITLE:      Generative Partial Multi-View Clustering
http://arxiv.org/abs/2003.13088
AUTHORS:        Qianqian Wang ; Zhengming Ding ; Zhiqiang Tao ; Quanxue Gao ; Yun Fu
COMMENTS:       This paper is an extension to our previous work: "Wang Q, Ding Z, Tao Z, et al. Partial multi-view clustering via consistent GAN[C]//2018 IEEE International Conference on Data Mining (ICDM). IEEE, 2018: 1290-1295."
HIGHLIGHT:      In this study, we design and build a generative partial multi-view clustering model, named as GP-MVC, to address the incomplete multi-view problem by explicitly generating the data of missing views.

80, TITLE:      Noise Modeling, Synthesis and Classification for Generic Object Anti-Spoofing
http://arxiv.org/abs/2003.13043
AUTHORS:        Joel Stehouwer ; Yaojie Liu ; Amin Jourabloo ; Xiaoming Liu
COMMENTS:       In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020
HIGHLIGHT:      In this work, we define and tackle the problem of Generic Object Anti-Spoofing (GOAS) for the first time.

81, TITLE:      Attentive CutMix: An Enhanced Data Augmentation Approach for Deep Learning Based Image Classification
http://arxiv.org/abs/2003.13048
AUTHORS:        Devesh Walawalkar ; Zhiqiang Shen ; Zechun Liu ; Marios Savvides
COMMENTS:       Accepted as conference paper in ICASSP 2020
HIGHLIGHT:      In this paper, we propose Attentive CutMix, a naturally enhanced augmentation strategy based on CutMix.

82, TITLE:      Deep Face Super-Resolution with Iterative Collaboration between Attentive Recovery and Landmark Estimation
http://arxiv.org/abs/2003.13063
AUTHORS:        Cheng Ma ; Zhenyu Jiang ; Yongming Rao ; Jiwen Lu ; Jie Zhou
COMMENTS:       Accepted to CVPR 2020
HIGHLIGHT:      In this paper, we propose a deep face super-resolution (FSR) method with iterative collaboration between two recurrent networks which focus on facial image recovery and landmark estimation respectively.

83, TITLE:      Omni-sourced Webly-supervised Learning for Video Recognition
http://arxiv.org/abs/2003.13042
AUTHORS:        Haodong Duan ; Yue Zhao ; Yuanjun Xiong ; Wentao Liu ; Dahua Lin
HIGHLIGHT:      We introduce OmniSource, a novel framework for leveraging web data to train video recognition models.

84, TITLE:      Disturbance-immune Weight Sharing for Neural Architecture Search
http://arxiv.org/abs/2003.13089
AUTHORS:        Shuaicheng Niu ; Jiaxiang Wu ; Yifan Zhang ; Yong Guo ; Peilin Zhao ; Junzhou Huang ; Mingkui Tan
HIGHLIGHT:      To alleviate the performance disturbance issue, we propose a new disturbance-immune update strategy for model updating.

85, TITLE:      Learning by Analogy: Reliable Supervision from Transformations for Unsupervised Optical Flow Estimation
http://arxiv.org/abs/2003.13045
AUTHORS:        Liang Liu ; Jiangning Zhang ; Ruifei He ; Yong Liu ; Yabiao Wang ; Ying Tai ; Donghao Luo ; Chengjie Wang ; Jilin Li ; Feiyue Huang
COMMENTS:       Accepted to CVPR 2020
HIGHLIGHT:      In this work, we present a framework to use more reliable supervision from transformations.

86, TITLE:      Source Printer Identification from Document Images Acquired using Smartphone
http://arxiv.org/abs/2003.12602
AUTHORS:        Sharad Joshi ; Suraj Saxena ; Nitin Khanna
COMMENTS:       10 pages
HIGHLIGHT:      Building on existing methods, we propose to learn a single CNN model from the fusion of letter images and their printer-specific noise residuals. In the absence of any publicly available dataset, we created a new dataset consisting of 2250 document images of text documents printed by eighteen printers and acquired by a smartphone camera at five acquisition settings.

87, TITLE:      On the Evaluation of Prohibited Item Classification and Detection in Volumetric 3D Computed Tomography Baggage Security Screening Imagery
http://arxiv.org/abs/2003.12625
AUTHORS:        Qian Wang ; Neelanjan Bhowmik ; Toby P. Breckon
COMMENTS:       Accepted to IJCNN 2020
HIGHLIGHT:      In this paper, we aim to evaluate the possibility of extending the automatic prohibited item detection from 2D X-ray imagery to volumetric 3D CT baggage security screening imagery.

88, TITLE:      SceneCAD: Predicting Object Alignments and Layouts in RGB-D Scans
http://arxiv.org/abs/2003.12622
AUTHORS:        Armen Avetisyan ; Tatiana Khanova ; Christopher Choy ; Denver Dash ; Angela Dai ; Matthias Nießner
COMMENTS:       Video here https://youtu.be/F0DpggYByh0
HIGHLIGHT:      We present a novel approach to reconstructing lightweight, CAD-based representations of scanned 3D environments from commodity RGB-D sensors.

89, TITLE:      Acceleration of Convolutional Neural Network Using FFT-Based Split Convolutions
http://arxiv.org/abs/2003.12621
AUTHORS:        Kamran Chitsaz ; Mohsen Hajabdollahi ; Nader Karimi ; Shadrokh Samavi ; Shahram Shirani
COMMENTS:       5 pages, 4 figures
HIGHLIGHT:      In this paper, a new method for CNN processing in the FFT domain is proposed, which is based on input splitting.

90, TITLE:      TREC CAsT 2019: The Conversational Assistance Track Overview
http://arxiv.org/abs/2003.13624
AUTHORS:        Jeffrey Dalton ; Chenyan Xiong ; Jamie Callan
HIGHLIGHT:      This year 21 groups submitted a total of 65 runs using varying methods for conversational query understanding and ranking.

91, TITLE:      AliCoCo: Alibaba E-commerce Cognitive Concept Net
http://arxiv.org/abs/2003.13230
AUTHORS:        Xusheng Luo ; Luxin Liu ; Yonghua Yang ; Le Bo ; Yuanpeng Cao ; Jinhang Wu ; Qiang Li ; Keping Yang ; Kenny Q. Zhu
COMMENTS:       15 pages. Accepted by SIGMOD 2020 Industry Track
HIGHLIGHT:      In this paper, we propose to construct a large-scale e-commerce cognitive concept net named "AliCoCo", which is practiced in Alibaba, the largest Chinese e-commerce platform in the world.

92, TITLE:      Learning Latent Causal Structures with a Redundant Input Neural Network
http://arxiv.org/abs/2003.13135
AUTHORS:        Jonathan D. Young ; Bryan Andrews ; Gregory F. Cooper ; Xinghua Lu
HIGHLIGHT:      In this paper, we address a problem for which it is known that inputs cause outputs, and these causal relationships are encoded by a causal network among a set of an unknown number of latent variables.

93, TITLE:      Environmental Adaptation of Robot Morphology and Control through Real-world Evolution
http://arxiv.org/abs/2003.13254
AUTHORS:        Tønnes F. Nygaard ; Charles P. Martin ; David Howard ; Jim Torresen ; Kyrre Glette
HIGHLIGHT:      In this paper, we rely solely on real-world evaluations and apply evolutionary search to yield combinations of morphology and control for our mechanically self-reconfiguring quadruped robot.

94, TITLE:      Image compression optimized for 3D reconstruction by utilizing deep neural networks
http://arxiv.org/abs/2003.12618
AUTHORS:        Alex Golts ; Yoav Y. Schechner
HIGHLIGHT:      Motivated by works on recurrent neural network (RNN)-based image compression and three-dimensional (3D) reconstruction, we propose unified network architectures to solve both tasks jointly.

95, TITLE:      European Language Grid: An Overview
http://arxiv.org/abs/2003.13551
AUTHORS:        Georg Rehm ; Maria Berger ; Ela Elsholz ; Stefanie Hegele ; Florian Kintzel ; Katrin Marheinecke ; Stelios Piperidis ; Miltos Deligiannis ; Dimitris Galanis ; Katerina Gkirtzou ; Penny Labropoulou ; Kalina Bontcheva ; David Jones ; Ian Roberts ; Jan Hajic ; Jana Hamrlová ; Lukáš Kačena ; Khalid Choukri ; Victoria Arranz ; Andrejs Vasiļjevs ; Orians Anvari ; Andis Lagzdiņš ; Jūlija Meļņika ; Gerhard Backfried ; Erinç Dikici ; Miroslav Janosik ; Katja Prinz ; Christoph Prinz ; Severin Stampler ; Dorothea Thomas-Aniola ; José Manuel Gómez Pérez ; Andres Garcia Silva ; Christian Berrío ; Ulrich Germann ; Steve Renals ; Ondrej Klejch
COMMENTS:       Proceedings of the 12th Language Resources and Evaluation Conference (LREC 2020). To appear
HIGHLIGHT:      The European Language Grid (ELG) project addresses this fragmentation by establishing the ELG as the primary platform for LT in Europe.

96, TITLE:      Mining Implicit Entity Preference from User-Item Interaction Data for Knowledge Graph Completion via Adversarial Learning
http://arxiv.org/abs/2003.12718
AUTHORS:        Gaole He ; Junyi Li ; Wayne Xin Zhao ; Peiju Liu ; Ji-Rong Wen
COMMENTS:       11 pages, 4 figures, 6 tables. Accepted as WWW 2020 paper
HIGHLIGHT:      In this paper, we take a new perspective that aims to leverage rich user-item interaction data (user interaction data for short) for improving the KGC task.

97, TITLE:      Countering Language Drift with Seeded Iterated Learning
http://arxiv.org/abs/2003.12694
AUTHORS:        Yuchen Lu ; Soumye Singhal ; Florian Strub ; Olivier Pietquin ; Aaron Courville
HIGHLIGHT:      In this paper, we propose a generic approach to counter language drift by using iterated learning.

98, TITLE:      QRMine: A python package for triangulation in Grounded Theory
http://arxiv.org/abs/2003.13519
AUTHORS:        Bell Raj Eapen ; Norm Archer ; Kamran Sartipi
HIGHLIGHT:      We present an open-source python package (QRMine) that encapsulates various ML and NLP libraries to support coding and triangulation in GT.

99, TITLE:      Adversarial Stress Testing of Lifetime Distributions
http://arxiv.org/abs/2003.12587
AUTHORS:        Nozer Singpurwalla
HIGHLIGHT:      In this paper we put forward the viewpoint that the notion of stress testing financial institutions and engineered systems can also be made viable appropos the stress testing an individual's strength of conviction in a probability distribution.

100, TITLE:     Super Resolution for Root Imaging
http://arxiv.org/abs/2003.13537
AUTHORS:        Jose F. Ruiz-Munoz ; Alina Zare ; Jyothier K. Nimmagadda ; Shuang Cui ; James E. Baciak
COMMENTS:       Under review. Submitted to Applications in Plant Sciences (APPS)
HIGHLIGHT:      In this study, we show that this issue can be tackled by the incorporation of a deep-learning based SR model in the image formation process.

101, TITLE:     Rethinking Depthwise Separable Convolutions: How Intra-Kernel Correlations Lead to Improved MobileNets
http://arxiv.org/abs/2003.13549
AUTHORS:        Daniel Haase ; Manuel Amthor
COMMENTS:       Accepted by CVPR 2020
HIGHLIGHT:      We introduce blueprint separable convolutions (BSConv) as highly efficient building blocks for CNNs.

102, TITLE:     Squeezed Deep 6DoF Object Detection Using Knowledge Distillation
http://arxiv.org/abs/2003.13586
AUTHORS:        Heitor Felix ; Walber M. Rodrigues ; David Macêdo ; Francisco Simões ; Adriano L. I. Oliveira ; Veronica Teichrieb ; Cleber Zanchettin
COMMENTS:       This paper was accepted by IJCNN 2020 and will have few changes from the version that will be published
HIGHLIGHT:      In this paper, we propose an approach to reduce the complexity of 6DoF detection networks while maintaining accuracy.

103, TITLE:     Laplacian Denoising Autoencoder
http://arxiv.org/abs/2003.13623
AUTHORS:        Jianbo Jiao ; Linchao Bao ; Yunchao Wei ; Shengfeng He ; Honghui Shi ; Rynson Lau ; Thomas S. Huang
HIGHLIGHT:      In this paper, we propose to learn data representations with a novel type of denoising autoencoder, where the noisy input data is generated by corrupting latent clean data in the gradient domain.

104, TITLE:     TResNet: High Performance GPU-Dedicated Architecture
http://arxiv.org/abs/2003.13630
AUTHORS:        Tal Ridnik ; Hussam Lawen ; Asaf Noy ; Itamar Friedman
COMMENTS:       9 pages, 5 figures
HIGHLIGHT:      In this work, we introduce a series of architecture modifications that aim to boost neural networks' accuracy, while retaining their GPU training and inference efficiency.

105, TITLE:     Speech2Action: Cross-modal Supervision for Action Recognition
http://arxiv.org/abs/2003.13594
AUTHORS:        Arsha Nagrani ; Chen Sun ; David Ross ; Rahul Sukthankar ; Cordelia Schmid ; Andrew Zisserman
COMMENTS:       Accepted to CVPR 2020
HIGHLIGHT:      In this work we investigate the link between spoken words and actions in movies.

106, TITLE:     Fast-MVSNet: Sparse-to-Dense Multi-View Stereo With Learned Propagation and Gauss-Newton Refinement
http://arxiv.org/abs/2003.13017
AUTHORS:        Zehao Yu ; Shenghua Gao
COMMENTS:       Accepted by CVPR2020
HIGHLIGHT:      Towards this end, this paper presents a Fast-MVSNet, a novel sparse-to-dense coarse-to-fine framework, for fast and accurate depth estimation in MVS.

107, TITLE:     Multi-Path Region Mining For Weakly Supervised 3D Semantic Segmentation on Point Clouds
http://arxiv.org/abs/2003.13035
AUTHORS:        Jiacheng Wei ; Guosheng Lin ; Kim-Hui Yap ; Tzu-Yi Hung ; Lihua Xie
COMMENTS:       Accepted by CVPR2020
HIGHLIGHT:      In this paper, we propose a weakly supervised approach to predict point-level results using weak labels on 3D point clouds.

108, TITLE:     Data-Driven Neuromorphic DRAM-based CNN and RNN Accelerators
http://arxiv.org/abs/2003.13006
AUTHORS:        Tobi Delbruck ; Shih-Chii Liu
COMMENTS:       To appear in 2019 IEEE Sig. Proc. Soc. Asilomar Conference on Signals, Systems, and Computers Session MP6b: Neuromorphic Computing (Invited)
HIGHLIGHT:      This paper reports on our developments over the last 5 years of convolutional and recurrent deep neural network hardware accelerators that exploit either spatial or temporal sparsity similar to SNNs but achieve SOA throughput, power efficiency and latency even with the use of DRAM for the required storage of the weights and states of large DNNs.

109, TITLE:     Self-Supervised Learning for Domain Adaptation on Point-Clouds
http://arxiv.org/abs/2003.12641
AUTHORS:        Idan Achituve ; Haggai Maron ; Gal Chechik
HIGHLIGHT:      We introduce a new pretext task, Region Reconstruction, motivated by the deformations encountered in sim-to-real transformation.

110, TITLE:     Combining Visible and Infrared Spectrum Imagery using Machine Learning for Small Unmanned Aerial System Detection
http://arxiv.org/abs/2003.12638
AUTHORS:        Vinicius G. Goecks ; Grayson Woods ; Niladri Das ; John Valasek
HIGHLIGHT:      This research work proposes combining the advantages of the LWIR and visible spectrum sensors using machine learning for vision-based detection of sUAS.

111, TITLE:     Deep 3D Capture: Geometry and Reflectance from Sparse Multi-View Images
http://arxiv.org/abs/2003.12642
AUTHORS:        Sai Bi ; Zexiang Xu ; Kalyan Sunkavalli ; David Kriegman ; Ravi Ramamoorthi
COMMENTS:       Accepted to CVPR 2020
HIGHLIGHT:      We introduce a novel learning-based method to reconstruct the high-quality geometry and complex, spatially-varying BRDF of an arbitrary object from a sparse set of only six images captured by wide-baseline cameras under collocated point lighting.

112, TITLE:     Designing Color Filters that Make Cameras MoreColorimetric
http://arxiv.org/abs/2003.12645
AUTHORS:        Graham D. Finlayson ; Yuteng Zhu
COMMENTS:       13 pages, 5 figures, 3 algorithms, journal
HIGHLIGHT:      In this paper, we solve for the filter which returns the modified sensitivities as close to being a linear transformation from the color matching functions of human visual system as possible.

113, TITLE:     Deep CG2Real: Synthetic-to-Real Translation via Image Disentanglement
http://arxiv.org/abs/2003.12649
AUTHORS:        Sai Bi ; Kalyan Sunkavalli ; Federico Perazzi ; Eli Shechtman ; Vladimir Kim ; Ravi Ramamoorthi
COMMENTS:       Accepted to ICCV 2019
HIGHLIGHT:      We present a method to improve the visual realism of low-quality, synthetic images, e.g. OpenGL renderings.

114, TITLE:     Detection and Description of Change in Visual Streams
http://arxiv.org/abs/2003.12633
AUTHORS:        Davis Gilton ; Ruotian Luo ; Rebecca Willett ; Greg Shakhnarovich
HIGHLIGHT:      This paper presents a framework for the analysis of changes in visual streams: ordered sequences of images, possibly separated by significant time gaps.

115, TITLE:     Policy Teaching via Environment Poisoning: Training-time Adversarial Attacks against Reinforcement Learning
http://arxiv.org/abs/2003.12909
AUTHORS:        Amin Rakhsha ; Goran Radanovic ; Rati Devidze ; Xiaojin Zhu ; Adish Singla
HIGHLIGHT:      We propose an optimization framework for finding an \emph{optimal stealthy attack} for different measures of attack cost.

116, TITLE:     Mutual Learning Network for Multi-Source Domain Adaptation
http://arxiv.org/abs/2003.12944
AUTHORS:        Zhenpeng Li ; Zhen Zhao ; Yuhong Guo ; Haifeng Shen ; Jieping Ye
HIGHLIGHT:      In this paper, we propose a novel multi-source domain adaptation method, Mutual Learning Network for Multiple Source Domain Adaptation (ML-MSDA).

117, TITLE:     A Benchmark for Point Clouds Registration Algorithms
http://arxiv.org/abs/2003.12841
AUTHORS:        Simone Fontana ; Daniele Cattaneo ; Augusto Luis Ballardini ; Matteo Vaghi ; Domenico Giorgio Sorrenti
HIGHLIGHT:      This work aims at encouraging authors to use a public and shared benchmark, instead than data collected ad-hoc, to ensure objectivity and repeatability, two fundamental characteristics in any scientific field. For these reasons, we developed a benchmark, for localization and mapping applications, using multiple publicly available datasets. Along with the data, we provide a broad set of registration problems, chosen to cover different types of initial misalignment, various degrees of overlap, and different kinds of registration problems.

118, TITLE:     Optimized Directed Roadmap Graph for Multi-Agent Path Finding Using Stochastic Gradient Descent
http://arxiv.org/abs/2003.12924
AUTHORS:        Christian Henkel ; Marc Toussaint
HIGHLIGHT:      We present a novel approach called Optimized Directed Roadmap Graph (ODRM).

119, TITLE:     Learning to Smooth and Fold Real Fabric Using Dense Object Descriptors Trained on Synthetic Color Images
http://arxiv.org/abs/2003.12698
AUTHORS:        Aditya Ganapathi ; Priya Sundaresan ; Brijen Thananjeyan ; Ashwin Balakrishna ; Daniel Seita ; Jennifer Grannen ; Minho Hwang ; Ryan Hoque ; Joseph E. Gonzalez ; Nawid Jamali ; Katsu Yamane ; Soshi Iba ; Ken Goldberg
HIGHLIGHT:      In this paper, we learn visual representations of deformable fabric by training dense object descriptors that capture correspondences across images of fabric in various configurations.

120, TITLE:     Fast Encoding of AG Codes over $C_{ab}$ Curves
http://arxiv.org/abs/2003.13333
AUTHORS:        Peter Beelen ; Johan Rosenkilde ; Grigory Solomatov
HIGHLIGHT:      We investigate algorithms for encoding of one-point algebraic geometry (AG) codes over certain plane curves called $C_{ab}$ curves, as well as algorithms for inverting the encoding map, which we call "unencoding".

121, TITLE:     Abstractive Text Summarization based on Language Model Conditioning and Locality Modeling
http://arxiv.org/abs/2003.13027
AUTHORS:        Dmitrii Aksenov ; Julián Moreno-Schneider ; Peter Bourgonje ; Robert Schwarzenberg ; Leonhard Hennig ; Georg Rehm
COMMENTS:       Proceedings of the 12th Language Resources and Evaluation Conference (LREC 2020). To appear
HIGHLIGHT:      We explore to what extent knowledge about the pre-trained language model that is used is beneficial for the task of abstractive summarization.

122, TITLE:     Meta Fine-Tuning Neural Language Models for Multi-Domain Text Mining
http://arxiv.org/abs/2003.13003
AUTHORS:        Chengyu Wang ; Minghui Qiu ; Jun Huang ; Xiaofeng He
HIGHLIGHT:      In this paper, we propose an effective learning procedure named Meta Fine-Tuning (MFT), served as a meta-learner to solve a group of similar NLP tasks for neural language models.

123, TITLE:     Abstractive Summarization with Combination of Pre-trained Sequence-to-Sequence and Saliency Models
http://arxiv.org/abs/2003.13028
AUTHORS:        Itsumi Saito ; Kyosuke Nishida ; Kosuke Nishida ; Junji Tomita
COMMENTS:       Work in progress
HIGHLIGHT:      In this study, we investigated the effectiveness of combining saliency models that identify the important parts of the source text with the pre-trained seq-to-seq models through extensive experiments.

124, TITLE:     Recursive Non-Autoregressive Graph-to-Graph Transformer for Dependency Parsing with Iterative Refinement
http://arxiv.org/abs/2003.13118
AUTHORS:        Alireza Mohammadshahi ; James Henderson
HIGHLIGHT:      We propose the Recursive Non-autoregressive Graph-to-graph Transformer architecture (RNG-Tr) for the iterative refinement of arbitrary graphs through the recursive application of a non-autoregressive Graph-to-Graph Transformer and apply it to syntactic dependency parsing.

125, TITLE:     A Dataset of German Legal Documents for Named Entity Recognition
http://arxiv.org/abs/2003.13016
AUTHORS:        Elena Leitner ; Georg Rehm ; Julián Moreno-Schneider
COMMENTS:       Proceedings of the 12th Language Resources and Evaluation Conference (LREC 2020). To appear
HIGHLIGHT:      We describe a dataset developed for Named Entity Recognition in German federal court decisions.

126, TITLE:     Named Entities in Medical Case Reports: Corpus and Experiments
http://arxiv.org/abs/2003.13032
AUTHORS:        Sarah Schulz ; Jurica Ševa ; Samuel Rodriguez ; Malte Ostendorff ; Georg Rehm
COMMENTS:       Proceedings of the 12th Language Resources and Evaluation Conference (LREC 2020). To appear
HIGHLIGHT:      In the case reports, we annotate cases, conditions, findings, factors and negation modifiers. We present a new corpus comprising annotations of medical entities in case reports, originating from PubMed Central's open access library.

127, TITLE:     Context Based Emotion Recognition using EMOTIC Dataset
http://arxiv.org/abs/2003.13401
AUTHORS:        Ronak Kosti ; Jose M. Alvarez ; Adria Recasens ; Agata Lapedriza
HIGHLIGHT:      In this paper we present EMOTIC, a dataset of images of people in a diverse set of natural situations, annotated with their apparent emotion.

128, TITLE:     Same Features, Different Day: Weakly Supervised Feature Learning for Seasonal Invariance
http://arxiv.org/abs/2003.13431
AUTHORS:        Jaime Spencer ; Richard Bowden ; Simon Hadfield
HIGHLIGHT:      The aim of this paper is to provide a dense feature representation that can be used to perform localization, sparse matching or image retrieval, regardless of the current seasonal or temporal appearance.

129, TITLE:     RPM-Net: Robust Point Matching using Learned Features
http://arxiv.org/abs/2003.13479
AUTHORS:        Zi Jian Yew ; Gim Hee Lee
COMMENTS:       10 pages, 4 figures. To appear in CVPR2020
HIGHLIGHT:      In this paper, we propose the RPM-Net -- a less sensitive to initialization and more robust deep learning-based approach for rigid point cloud registration.

130, TITLE:     Predicting Semantic Map Representations from Images using Pyramid Occupancy Networks
http://arxiv.org/abs/2003.13402
AUTHORS:        Thomas Roddick ; Roberto Cipolla
HIGHLIGHT:      In this work we present a simple, unified approach for estimating maps directly from monocular images using a single end-to-end deep learning architecture.

131, TITLE:     DeFeat-Net: General Monocular Depth via Simultaneous Unsupervised Representation Learning
http://arxiv.org/abs/2003.13446
AUTHORS:        Jaime Spencer ; Richard Bowden ; Simon Hadfield
HIGHLIGHT:      In the current monocular depth research, the dominant approach is to employ unsupervised training on large datasets, driven by warped photometric consistency.

132, TITLE:     AutoTrack: Towards High-Performance Visual Tracking for UAV with Automatic Spatio-Temporal Regularization
http://arxiv.org/abs/2003.12949
AUTHORS:        Yiming Li ; Changhong Fu ; Fangqiang Ding ; Ziyuan Huang ; Geng Lu
COMMENTS:       2020 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
HIGHLIGHT:      In this work, a novel approach is proposed to online automatically and adaptively learn spatio-temporal regularization term.

133, TITLE:     Superpixel Segmentation with Fully Convolutional Networks
http://arxiv.org/abs/2003.12929
AUTHORS:        Fengting Yang ; Qian Sun ; Hailin Jin ; Zihan Zhou
COMMENTS:       16 pages, 15 figures, to be published in CVPR'20
HIGHLIGHT:      Inspired by an initialization strategy commonly adopted by traditional superpixel algorithms, we present a novel method that employs a simple fully convolutional network to predict superpixels on a regular image grid.

134, TITLE:     Co-occurrence Background Model with Superpixels for Robust Background Initialization
http://arxiv.org/abs/2003.12931
AUTHORS:        Wenjun Zhou ; Yuheng Deng ; Bo Peng ; Dong Liang ; Shun'ichi Kaneko
HIGHLIGHT:      Background initialization is an important step in many high-level applications of video processing,ranging from video surveillance to video inpainting.However,this process is often affected by practical challenges such as illumination changes,background motion,camera jitter and intermittent movement,etc.In this paper,we develop a co-occurrence background model with superpixel segmentation for robust background initialization.

135, TITLE:     Adaptive Object Detection with Dual Multi-Label Prediction
http://arxiv.org/abs/2003.12943
AUTHORS:        Zhen Zhao ; Yuhong Guo ; Haifeng Shen ; Jieping Ye
HIGHLIGHT:      In this paper, we propose a novel end-to-end unsupervised deep domain adaptation model for adaptive object detection by exploiting multi-label object recognition as a dual auxiliary task.

136, TITLE:     A Novel Method of Extracting Topological Features from Word Embeddings
http://arxiv.org/abs/2003.13074
AUTHORS:        Shafie Gholizadeh ; Armin Seyeditabari ; Wlodek Zadrozny
HIGHLIGHT:      In this paper, we introduce a novel algorithm to extract topological features from word embedding representation of text that can be used for text classification.

137, TITLE:     Optimizing Geometry Compression using Quantum Annealing
http://arxiv.org/abs/2003.13253
AUTHORS:        Sebastian Feld ; Markus Friedrich ; Claudia Linnhoff-Popien
COMMENTS:       6 pages, 3 figures
HIGHLIGHT:      We propose a quantum-enabled lossy 3d point cloud compression pipeline based on the constructive solid geometry (CSG) model representation.

138, TITLE:     InterBERT: Vision-and-Language Interaction for Multi-modal Pretraining
http://arxiv.org/abs/2003.13198
AUTHORS:        Junyang Lin ; An Yang ; Yichang Zhang ; Jie Liu ; Jingren Zhou ; Hongxia Yang
COMMENTS:       11 pages, 4 figures
HIGHLIGHT:      In this work, we propose a novel model, namely InterBERT (BERT for Interaction), which owns strong capability of modeling interaction between the information flows of different modalities. Besides, we propose a large-scale dataset for multi-modal pretraining in Chinese, and we develop the Chinese InterBERT which is the first Chinese multi-modal pretrained model.

139, TITLE:     How human judgment impairs automated deception detection performance
http://arxiv.org/abs/2003.13316
AUTHORS:        Bennett Kleinberg ; Bruno Verschuere
HIGHLIGHT:      Method: We collected a corpus of truthful and deceptive answers about participants' autobiographical intentions (n=1640) and tested whether a combination of supervised machine learning and human judgment could improve deception detection accuracy.

140, TITLE:     Investigating Language Impact in Bilingual Approaches for Computational Language Documentation
http://arxiv.org/abs/2003.13325
AUTHORS:        Marcely Zanon Boito ; Aline Villavicencio ; Laurent Besacier
COMMENTS:       Accepted to 1st Joint SLTU and CCURL Workshop
HIGHLIGHT:      In this paper we investigate how the choice of translation language affects the posterior documentation work and potential automatic approaches which will work on top of the produced bilingual corpus.

141, TITLE:     A Corpus of Controlled Opinionated and Knowledgeable Movie Discussions for Training Neural Conversation Models
http://arxiv.org/abs/2003.13342
AUTHORS:        Fabian Galetzka ; Chukwuemeka U. Eneh ; David Schlangen
COMMENTS:       8 Pages, 8 Figures, 5 Tables. Accepted paper for LREC 2020 conference
HIGHLIGHT:      To address this, we introduce a new labeled dialogue dataset in the domain of movie discussions, where every dialogue is based on pre-specified facts and opinions.

142, TITLE:     Making Metadata Fit for Next Generation Language Technology Platforms: The Metadata Schema of the European Language Grid
http://arxiv.org/abs/2003.13236
AUTHORS:        Penny Labropoulou ; Katerina Gkirtzou ; Maria Gavriilidou ; Miltos Deligiannis ; Dimitrios Galanis ; Stelios Piperidis ; Georg Rehm ; Maria Berger ; Valérie Mapelli ; Mickaël Rigault ; Victoria Arranz ; Khalid Choukri ; Gerhard Backfried ; José Manuel Gómez Pérez ; Andres Garcia Silva
COMMENTS:       Proceedings of the 12th Language Resources and Evaluation Conference (LREC 2020). To appear
HIGHLIGHT:      In this paper we present ELG-SHARE, a rich metadata schema catering for the description of Language Resources and Technologies (processing and generation services and tools, models, corpora, term lists, etc.), as well as related entities (e.g., organizations, projects, supporting documents, etc.).

143, TITLE:     Learning Contextualized Sentence Representations for Document-Level Neural Machine Translation
http://arxiv.org/abs/2003.13205
AUTHORS:        Pei Zhang ; Xu Zhang ; Wei Chen ; Jian Yu ; Yanfeng Wang ; Deyi Xiong
HIGHLIGHT:      In this paper, we propose a new framework to model cross-sentence dependencies by training neural machine translation (NMT) to predict both the target translation and surrounding sentences of a source sentence.

144, TITLE:     Empirical Analysis of Zipf's Law, Power Law, and Lognormal Distributions in Medical Discharge Reports
http://arxiv.org/abs/2003.13352
AUTHORS:        Juan C Quiroz ; Liliana Laranjo ; Catalin Tufanaru ; Ahmet Baki Kocaballi ; Dana Rezazadegan ; Shlomo Berkovsky ; Enrico Coiera
HIGHLIGHT:      We examined 20,000 medical discharge reports from the MIMIC-III dataset.

145, TITLE:     Towards Palmprint Verification On Smartphones
http://arxiv.org/abs/2003.13266
AUTHORS:        Yingyi Zhang ; Lin Zhang ; Ruixin Zhang ; Shaoxin Li ; Jilin Li ; Feiyue Huang
HIGHLIGHT:      In this paper, aiming to fill the aforementioned research gap, we conducted a thorough study of palmprint verification on smartphones and our contributions are twofold.

146, TITLE:     Unsupervised Model Personalization while Preserving Privacy and Scalability: An Open Problem
http://arxiv.org/abs/2003.13296
AUTHORS:        Matthias De Lange ; Xu Jia ; Sarah Parisot ; Ales Leonardis ; Gregory Slabaugh ; Tinne Tuytelaars
COMMENTS:       CVPR 2020
HIGHLIGHT:      We aim to address this challenge within the continual learning paradigm and provide a novel Dual User-Adaptation framework (DUA) to explore the problem.

147, TITLE:     Multi-Objective Matrix Normalization for Fine-grained Visual Recognition
http://arxiv.org/abs/2003.13272
AUTHORS:        Shaobo Min ; Hantao Yao ; Hongtao Xie ; Zheng-Jun Zha ; Yongdong Zhang
HIGHLIGHT:      In this paper, we propose an efficient Multi-Objective Matrix Normalization (MOMN) method that can simultaneously normalize a bilinear representation in terms of square-root, low-rank, and sparsity.

148, TITLE:     Architecture Disentanglement for Deep Neural Networks
http://arxiv.org/abs/2003.13268
AUTHORS:        Jie Hu ; Rongrong Ji ; Qixiang Ye ; Tong Tong ; ShengChuan Zhang ; Ke Li ; Feiyue Huang ; Ling Shao
HIGHLIGHT:      To address this issue, we introduce the novel concept of Neural Architecture Disentanglement (NAD) in this paper.

149, TITLE:     PANDA: Prototypical Unsupervised Domain Adaptation
http://arxiv.org/abs/2003.13274
AUTHORS:        Dapeng Hu ; Jian Liang ; Qibin Hou ; Hanshu Yan ; Yunpeng Chen ; Shuicheng Yan ; Jiashi Feng
HIGHLIGHT:      In this work, we attempt to calibrate the noisy pseudo labels with prototypes.

150, TITLE:     Strip Pooling: Rethinking Spatial Pooling for Scene Parsing
http://arxiv.org/abs/2003.13328
AUTHORS:        Qibin Hou ; Li Zhang ; Ming-Ming Cheng ; Jiashi Feng
COMMENTS:       Published as a CVPR2020 paper
HIGHLIGHT:      In this paper, beyond conventional spatial pooling that usually has a regular shape of NxN, we rethink the formulation of spatial pooling by introducing a new pooling strategy, called strip pooling, which considers a long but narrow kernel, i.e., 1xN or Nx1.

151, TITLE:     Active stereo vision three-dimensional reconstruction by RGB dot pattern projection and ray intersection
http://arxiv.org/abs/2003.13322
AUTHORS:        Yongcan Shuang ; Zhenzhou Wang
HIGHLIGHT:      In this paper, we propose a new pattern extraction method and a new stereo vision matching method based on our novel structured light pattern.

152, TITLE:     Real-time Fruit Recognition and Grasp Estimation for Autonomous Apple harvesting
http://arxiv.org/abs/2003.13298
AUTHORS:        Hanwen Kang ; Chao Chen
HIGHLIGHT:      In this research, a fully neural network based visual perception framework for autonomous apple harvesting is proposed.

153, TITLE:     CNN-based Density Estimation and Crowd Counting: A Survey
http://arxiv.org/abs/2003.12783
AUTHORS:        Guangshuai Gao ; Junyu Gao ; Qingjie Liu ; Qi Wang ; Yunhong Wang
HIGHLIGHT:      In this paper, we have surveyed over 220 works to comprehensively and systematically study the crowd counting models, mainly CNN-based density map estimation methods.

154, TITLE:     CAKES: Channel-wise Automatic KErnel Shrinking for Efficient 3D Network
http://arxiv.org/abs/2003.12798
AUTHORS:        Qihang Yu ; Yingwei Li ; Jieru Mei ; Yuyin Zhou ; Alan L. Yuille
HIGHLIGHT:      In this paper, we propose Channel-wise Automatic KErnel Shrinking (CAKES), to enable efficient 3D learning by shrinking standard 3D convolutions into a set of economic operations (e.g., 1D, 2D convolutions).

155, TITLE:     Real-MFF Dataset: A Large Realistic Multi-focus Image Dataset with Ground Truth
http://arxiv.org/abs/2003.12779
AUTHORS:        Juncheng Zhang ; Qingmin Liao ; Shaojun Liu ; Haoyu Ma ; Wenming Yang ; Jing-hao Xue
HIGHLIGHT:      In this paper, we introduce a large and realistic multi-focus dataset containing 800 pairs of source images with the corresponding ground truth images.

156, TITLE:     Polarized Reflection Removal with Perfect Alignment in the Wild
http://arxiv.org/abs/2003.12789
AUTHORS:        Chenyang Lei ; Xuhua Huang ; Mengdi Zhang ; Qiong Yan ; Wenxiu Sun ; Qifeng Chen
COMMENTS:       CVPR2020
HIGHLIGHT:      We present a novel formulation to removing reflection from polarized images in the wild. Then we build a new dataset with more than 100 types of glass in which obtained transmission images are perfectly aligned with input mixed images.

157, TITLE:     Multi-Task Reinforcement Learning with Soft Modularization
http://arxiv.org/abs/2003.13661
AUTHORS:        Ruihan Yang ; Huazhe Xu ; Yi Wu ; Xiaolong Wang
COMMENTS:       Our project page: https://rchalyang.github.io/SoftModule
HIGHLIGHT:      Thus, instead of naively sharing parameters across tasks, we introduce an explicit modularization technique on policy representation to alleviate this optimization issue.

158, TITLE:     How Not to Give a FLOP: Combining Regularization and Pruning for Efficient Inference
http://arxiv.org/abs/2003.13593
AUTHORS:        Tai Vu ; Emily Wen ; Roy Nehoran
HIGHLIGHT:      In this paper, we examine the use of both regularization and pruning for reduced computational complexity and more efficient inference in Deep Neural Networks (DNNs).

159, TITLE:     The Hessian Estimation Evolution Strategy
http://arxiv.org/abs/2003.13256
AUTHORS:        Tobias Glasmachers ; Oswin Krause
HIGHLIGHT:      We present a novel black box optimization algorithm called Hessian Estimation Evolution Strategy.

160, TITLE:     Templates and Recurrences: Better Together
http://arxiv.org/abs/2003.13515
AUTHORS:        Jason Breck ; John Cyphert ; Zachary Kincaid ; Thomas Reps
COMMENTS:       20 pages, 3 figures
HIGHLIGHT:      In this paper, we combine these two approaches and obtain a technique that uses templates in which the unknowns are functions rather than numbers, and the constraints on the unknowns are recurrences.

161, TITLE:     SHX: Search History Driven Crossover for Real-Coded Genetic Algorithm
http://arxiv.org/abs/2003.13508
AUTHORS:        Takumi Nakane ; Xuequan Lu ; Chao Zhang
COMMENTS:       GECCO2020 poster
HIGHLIGHT:      To boost the performance of crossover in real-coded genetic algorithm (RCGA), in this paper we propose to exploit the search history cached so far in an online style during the iteration.

162, TITLE:     Distributed Embodied Evolution in Networks of Agents
http://arxiv.org/abs/2003.12848
AUTHORS:        Anil Yaman ; Giovanni Iacca
HIGHLIGHT:      In this work we propose a distributed embodied evolutionary approach to optimize spatially distributed, locally interacting agents by allowing them to exchange their behavior parameters and learn from each other to adapt to a certain task within a given environment.

163, TITLE:     First-order Gradual Information Flow Types with Gradual Guarantees
http://arxiv.org/abs/2003.12819
AUTHORS:        Abhishek Bichhawat ; McKenna McCall ; Limin Jia
HIGHLIGHT:      In this paper, we re-examine the connection between gradual information flow types and information flow monitors, and identify the root cause for the tension between satisfying gradual guarantees and noninterference.

164, TITLE:     Critical Limits in a Bump Attractor Network of Spiking Neurons
http://arxiv.org/abs/2003.13365
AUTHORS:        Alberto Arturo Vergani ; Christian Robert Huyck
HIGHLIGHT:      Since the bump network could behave in many ways, this paper explores some critical limits of the parameter space using various positive and negative weights and an increasing size of the input spike sources The neuromorphic simulation of the bumpattractor network shows that it exhibits a stationary, a splitting and a divergent spike pattern, in relation to different sets of weights and input windows.

165, TITLE:     Weakly-supervised land classification for coastal zone based on deep convolutional neural networks by incorporating dual-polarimetric characteristics into training dataset
http://arxiv.org/abs/2003.13648
AUTHORS:        Sheng Sun ; Armando Marino ; Wenze Shui ; Zhongwen Hu
HIGHLIGHT:      In this work we explore the performance of DCNNs on semantic segmentation using spaceborne polarimetric synthetic aperture radar (PolSAR) datasets.

166, TITLE:     BVI-DVC: A Training Database for Deep Video Compression
http://arxiv.org/abs/2003.13552
AUTHORS:        Di Ma ; Fan Zhang ; David R. Bull
HIGHLIGHT:      In this paper, a new extensive and representative video database, BVI-DVC, is presented for training CNN-based coding tools.



==========Updates to Previous Papers==========
1, TITLE:       A Robotic 3D Perception System for Operating Room Environment Awareness
http://arxiv.org/abs/2003.09487
AUTHORS:        Zhaoshuo Li ; Amirreza Shaban ; Jean-Gabriel Simard ; Dinesh Rabindran ; Simon DiMaio ; Omid Mohareri
COMMENTS:       Accepted in IPCAI 2020
HIGHLIGHT:      Purpose: We describe a 3D multi-view perception system for the da Vinci surgical system to enable Operating room (OR) scene understanding and context awareness.

2, TITLE:       A CNN-RNN Framework for Image Annotation from Visual Cues and Social Network Metadata
http://arxiv.org/abs/1910.05770
AUTHORS:        Tobia Tesan ; Pasquale Coscia ; Lamberto Ballan
HIGHLIGHT:      Our models use multiple semantic embeddings to achieve the dual objective of being robust to vocabulary changes between train and test sets and decoupling the architecture from the low-level metadata representation.

3, TITLE:       G$^{3}$AN: Disentangling Appearance and Motion for Video Generation
http://arxiv.org/abs/1912.05523
AUTHORS:        Yaohui Wang ; Piotr Bilinski ; Francois Bremond ; Antitza Dantcheva
COMMENTS:       CVPR 2020
HIGHLIGHT:      To tackle this challenge, we introduce G$^{3}$AN, a novel spatio-temporal generative model, which seeks to capture the distribution of high dimensional video data and to model appearance and motion in disentangled manner.

4, TITLE:       Fast(er) Reconstruction of Shredded Text Documents via Self-Supervised Deep Asymmetric Metric Learning
http://arxiv.org/abs/2003.10063
AUTHORS:        Thiago M. Paixão ; Rodrigo F. Berriel ; Maria C. S. Boeres ; Alessando L. Koerich ; Claudine Badue ; Alberto F. De Souza ; Thiago Oliveira-Santos
COMMENTS:       Accepted to CVPR 2020. Main Paper (9 pages, 10 figures) and Supplementary Material (5 pages, 9 figures)
HIGHLIGHT:      This work proposes a scalable deep learning approach for measuring pairwise compatibility in which the number of inferences scales linearly (rather than quadratically) with the number of shreds.

5, TITLE:       A Topological Nomenclature for 3D Shape Analysis in Connectomics
http://arxiv.org/abs/1909.12887
AUTHORS:        Abhimanyu Talwar ; Zudi Lin ; Donglai Wei ; Yuesong Wu ; Bowen Zheng ; Jinglin Zhao ; Won-Dong Jang ; Xueying Wang ; Jeff W. Lichtman ; Hanspeter Pfister
COMMENTS:       Technical report
HIGHLIGHT:      In this paper, we develop a novel topological nomenclature system to name these objects like the appellation for chemical compounds to promote neuroscience analysis based on their skeletal structures. To advance neuroscience, we will release a 3D segmentation dataset of mitochondria and pyramidal neurons reconstructed from a 100um cube electron microscopy volume with their reduced graph and topological nomenclature annotations.

6, TITLE:       Local Class-Specific and Global Image-Level Generative Adversarial Networks for Semantic-Guided Scene Generation
http://arxiv.org/abs/1912.12215
AUTHORS:        Hao Tang ; Dan Xu ; Yan Yan ; Philip H. S. Torr ; Nicu Sebe
COMMENTS:       Accepted to CVPR 2020, camera ready (10 pages) + supplementary (18 pages)
HIGHLIGHT:      In this paper, we address the task of semantic-guided scene generation.

7, TITLE:       Compression of descriptor models for mobile applications
http://arxiv.org/abs/2001.03102
AUTHORS:        Roy Miles ; Krystian Mikolajczyk
HIGHLIGHT:      To resolve this, we propose the Convolution-Depthwise-Pointwise(CDP) layer, which provides a means of interpolating between the standard and depthwise separable convolutions.

8, TITLE:       clDice -- a Topology-Preserving Loss Function for Tubular Structure Segmentation
http://arxiv.org/abs/2003.07311
AUTHORS:        Suprosanna Shit ; Johannes C. Paetzold ; Anjany Sekuboyina ; Andrey Zhylka ; Ivan Ezhov ; Alexander Unger ; Josien P. W. Pluim ; Giles Tetteh ; Bjoern H. Menze
COMMENTS:       * The authors Suprosanna Shit and Johannes C. Paetzold contributed equally to the work
HIGHLIGHT:      We introduce a novel similarity measure termed clDice, which is calculated on the intersection of the segmentation masks and their (morphological) skeletons.

9, TITLE:       A Quantum Computational Approach to Correspondence Problems on Point Sets
http://arxiv.org/abs/1912.12296
AUTHORS:        Vladislav Golyanik ; Christian Theobalt
COMMENTS:       11 pages, 5 figures, 2 tables, CVPR
HIGHLIGHT:      We review AQC and derive a new algorithm for correspondence problems on point sets suitable for execution on AQC.

10, TITLE:      DCDLearn: Multi-order Deep Cross-distance Learning for Vehicle Re-Identification
http://arxiv.org/abs/2003.11315
AUTHORS:        Rixing Zhu ; Jianwu Fang ; Hongke Xu ; Hongkai Yu ; Jianru Xue
HIGHLIGHT:      Specially, we treat the transferred images and the reconstructed images generated by one-view CycleGAN as multi-order augmented data for deep cross-distance learning, where the cross distances of multi-order image set with distinct identities are learned by optimizing an objective function with multi-order augmented triplet loss and center loss to achieve the camera-invariance and identity-consistency.

11, TITLE:      You2Me: Inferring Body Pose in Egocentric Video via First and Second Person Interactions
http://arxiv.org/abs/1904.09882
AUTHORS:        Evonne Ng ; Donglai Xiang ; Hanbyul Joo ; Kristen Grauman
HIGHLIGHT:      We propose a learning-based approach to estimate the camera wearer's 3D body pose from egocentric video sequences.

12, TITLE:      Listen to Look: Action Recognition by Previewing Audio
http://arxiv.org/abs/1912.04487
AUTHORS:        Ruohan Gao ; Tae-Hyun Oh ; Kristen Grauman ; Lorenzo Torresani
COMMENTS:       Appears in CVPR 2020; Project page: http://vision.cs.utexas.edu/projects/listen_to_look/
HIGHLIGHT:      We propose a framework for efficient action recognition in untrimmed video that uses audio as a preview mechanism to eliminate both short-term and long-term visual redundancies.

13, TITLE:      Infant brain MRI segmentation with dilated convolution pyramid downsampling and self-attention
http://arxiv.org/abs/1912.12570
AUTHORS:        Zhihao Lei ; Lin Qi ; Ying Wei ; Yunlong Zhou
COMMENTS:       5 pages,5 figures
HIGHLIGHT:      In this paper, we propose a dual aggregation network to adaptively aggregate different information in infant brain MRI segmentation.

14, TITLE:      A Simple Framework for Contrastive Learning of Visual Representations
http://arxiv.org/abs/2002.05709
AUTHORS:        Ting Chen ; Simon Kornblith ; Mohammad Norouzi ; Geoffrey Hinton
COMMENTS:       code and pretrained models at https://github.com/google-research/simclr
HIGHLIGHT:      This paper presents SimCLR: a simple framework for contrastive learning of visual representations.

15, TITLE:      Deep Reinforcement Learning with Weighted Q-Learning
http://arxiv.org/abs/2003.09280
AUTHORS:        Andrea Cini ; Carlo D'Eramo ; Jan Peters ; Cesare Alippi
COMMENTS:       Corrected typos
HIGHLIGHT:      In this work, we provide the methodological advances to benefit from the WQL properties in Deep Reinforcement Learning (DRL), by using neural networks with Dropout Variational Inference as an effective approximation of deep Gaussian processes.

16, TITLE:      Preventing the Generation of Inconsistent Sets of Classification Rules
http://arxiv.org/abs/1908.09652
AUTHORS:        Thiago Zafalon Miranda ; Diorge Brognara Sardinha ; Ricardo Cerri
HIGHLIGHT:      In this work, we propose two algorithms capable of finding feature-space regions inside which any created rule would be consistent with the already existing rules, preventing inconsistencies from arising.

17, TITLE:      Massive vs. Curated Word Embeddings for Low-Resourced Languages. The Case of Yorùbá and Twi
http://arxiv.org/abs/1912.02481
AUTHORS:        Jesujoba O. Alabi ; Kwabena Amponsah-Kaakyire ; David I. Adelani ; Cristina España-Bonet
COMMENTS:       9 pages, 4 tables. Accepted at LREC 2020
HIGHLIGHT:      In this paper we focus on two African languages, Yor\`ub\'a and Twi, and compare the word embeddings obtained in this way, with word embeddings obtained from curated corpora and a language-dependent processing. We analyse the noise in the publicly available corpora, collect high quality and noisy data for the two languages and quantify the improvements that depend not only on the amount of data but on the quality too. As output of the work, we provide corpora, embeddings and the test suits for both languages.

18, TITLE:      Cross-Lingual Adaptation Using Universal Dependencies
http://arxiv.org/abs/2003.10816
AUTHORS:        Nasrin Taghizadeh ; Heshaam Faili
HIGHLIGHT:      We describe a cross-lingual adaptation method based on syntactic parse trees obtained from the Universal Dependencies (UD), which are consistent across languages, to develop classifiers in low-resource languages.

19, TITLE:      CPR-GCN: Conditional Partial-Residual Graph Convolutional Network in Automated Anatomical Labeling of Coronary Arteries
http://arxiv.org/abs/2003.08560
AUTHORS:        Han Yang ; Xingjian Zhen ; Ying Chi ; Lei Zhang ; Xian-Sheng Hua
COMMENTS:       This work is done by Xingjian Zhen during internship in Alibaba Damo Academy
HIGHLIGHT:      Motivated by the wide application of the graph neural network in structured data, in this paper, we propose a conditional partial-residual graph convolutional network (CPR-GCN), which takes both position and CT image into consideration, since CT image contains abundant information such as branch size and spanning direction.

20, TITLE:      Li Yuan, Tao Wang, Xiaopeng Zhang, Francis EH Tay, Zequn Jie, Wei Liu, Jiashi Feng
http://arxiv.org/abs/1908.00347
AUTHORS:        Li Yuan ; Tao Wang ; Xiaopeng Zhang ; Zequn Jie ; Francis EH Tay ; Jiashi Feng
COMMENTS:       CVPR2020, Codes: https://github.com/yuanli2333/Hadamard-Matrix-for-hashing
HIGHLIGHT:      In this work, we propose a new \emph{global} similarity metric, termed as \emph{central similarity}, with which the hash codes of similar data pairs are encouraged to approach a common center and those for dissimilar pairs to converge to different centers, to improve hash learning efficiency and retrieval accuracy.

21, TITLE:      Towards Model-Agnostic Adversarial Defenses using Adversarially Trained Autoencoders
http://arxiv.org/abs/1909.05921
AUTHORS:        Pratik Vaishnavi ; Kevin Eykholt ; Atul Prakash ; Amir Rahmati
HIGHLIGHT:      In this work, we take a first step toward creating a model-agnostic adversarial defense.

22, TITLE:      Detecting Deficient Coverage in Colonoscopies
http://arxiv.org/abs/2001.08589
AUTHORS:        Daniel Freedman ; Yochai Blau ; Liran Katzir ; Amit Aides ; Ilan Shimshoni ; Danny Veikherman ; Tomer Golany ; Ariel Gordon ; Greg Corrado ; Yossi Matias ; Ehud Rivlin
HIGHLIGHT:      This paper attempts to rectify the problem of substandard coverage in colonoscopy through the introduction of the C2D2 (Colonoscopy Coverage Deficiency via Depth) algorithm which detects deficient coverage, and can thereby alert the endoscopist to revisit a given area.

23, TITLE:      LAMP-HQ: A Large-Scale Multi-Pose High-Quality Database and Benchmark for NIR-VIS Face Recognition
http://arxiv.org/abs/1912.07809
AUTHORS:        Aijing Yu ; Haoxue Wu ; Huaibo Huang ; Zhen Lei ; Ran He
HIGHLIGHT:      In this paper, we propose a new Large-Scale Multi-Pose High-Quality NIR-VIS database LAMP-HQ containing 56,788 NIR and 16,828 VIS images of 573 subjects with large diversities in pose, illumination, attribute, scene and accessory.

24, TITLE:      Open Compound Domain Adaptation
http://arxiv.org/abs/1909.03403
AUTHORS:        Ziwei Liu ; Zhongqi Miao ; Xingang Pan ; Xiaohang Zhan ; Dahua Lin ; Stella X. Yu ; Boqing Gong
COMMENTS:       To appear in CVPR 2020 as an oral presentation. Code, datasets and models are available at: https://liuziwei7.github.io/projects/CompoundDomain.html
HIGHLIGHT:      We propose a new approach based on two technical insights into OCDA: 1) a curriculum domain adaptation strategy to bootstrap generalization across domains in a data-driven self-organizing fashion and 2) a memory module to increase the model's agility towards novel domains.

25, TITLE:      Show, Match and Segment: Joint Weakly Supervised Learning of Semantic Matching and Object Co-segmentation
http://arxiv.org/abs/1906.05857
AUTHORS:        Yun-Chun Chen ; Yen-Yu Lin ; Ming-Hsuan Yang ; Jia-Bin Huang
COMMENTS:       PAMI 2020. Project: https://yunchunchen.github.io/MaCoSNet-web/ Code: https://github.com/YunChunChen/MaCoSNet-pytorch
HIGHLIGHT:      We present an approach for jointly matching and segmenting object instances of the same category within a collection of images.

26, TITLE:      Tight Polynomial Worst-Case Bounds for Loop Programs
http://arxiv.org/abs/1906.10047
AUTHORS:        Amir M. Ben-Amram ; Geoff Hamilton
COMMENTS:       40 pages
HIGHLIGHT:      This paper shows how to obtain asymptotically-tight, multivariate, disjunctive polynomial bounds for this class of programs.

27, TITLE:      Towards Logical Specification of Statistical Machine Learning
http://arxiv.org/abs/1907.10327
AUTHORS:        Yusuke Kawamoto
COMMENTS:       SEFM'19 conference paper (full version with errors corrected)
HIGHLIGHT:      We introduce a logical approach to formalizing statistical properties of machine learning.

28, TITLE:      Invariant Risk Minimization
http://arxiv.org/abs/1907.02893
AUTHORS:        Martin Arjovsky ; Léon Bottou ; Ishaan Gulrajani ; David Lopez-Paz
HIGHLIGHT:      We introduce Invariant Risk Minimization (IRM), a learning paradigm to estimate invariant correlations across multiple training distributions.

29, TITLE:      An Online-Learning Approach to Inverse Optimization
http://arxiv.org/abs/1810.12997
AUTHORS:        Andreas Bärmann ; Alexander Martin ; Sebastian Pokutta ; Oskar Schneider
HIGHLIGHT:      In this paper, we demonstrate how to learn the objective function of a decision-maker while only observing the problem input data and the decision-maker's corresponding decisions over multiple rounds.

30, TITLE:      $P\neq NP$
http://arxiv.org/abs/2003.09791
AUTHORS:        Tianrong Lin
COMMENTS:       Submission for review, comments are welcome (v2:the proof of lemma3.1 and lemma3.3 deleted; v3: the claimed proof of theorem1.4 and theorem 1.6 improved)
HIGHLIGHT:      The main contribution of the present paper is that a series of results are obtained.

31, TITLE:      Unsupervised Question Decomposition for Question Answering
http://arxiv.org/abs/2002.09758
AUTHORS:        Ethan Perez ; Patrick Lewis ; Wen-tau Yih ; Kyunghyun Cho ; Douwe Kiela
COMMENTS:       Code available at https://github.com/facebookresearch/UnsupervisedDecomposition
HIGHLIGHT:      Since collecting labeled decompositions is cumbersome, we propose an unsupervised approach to produce sub-questions.

32, TITLE:      Lung Infection Quantification of COVID-19 in CT Images with Deep Learning
http://arxiv.org/abs/2003.04655
AUTHORS:        Fei Shan ; Yaozong Gao ; Jun Wang ; Weiya Shi ; Nannan Shi ; Miaofei Han ; Zhong Xue ; Dinggang Shen ; Yuxin Shi
COMMENTS:       23 pages, 6 figures
HIGHLIGHT:      In this paper, a deep learning (DL)-based segmentation system is developed to automatically quantify infection regions of interest (ROIs) and their volumetric ratios w.r.t. the lung.

33, TITLE:      Deep Line Art Video Colorization with a Few References
http://arxiv.org/abs/2003.10685
AUTHORS:        Min Shi ; Jia-Qi Zhang ; Shu-Yu Chen ; Lin Gao ; Yu-Kun Lai ; Fang-Lue Zhang
HIGHLIGHT:      In this paper, we propose a deep architecture to automatically color line art videos with the same color style as the given reference images.

34, TITLE:      Fixing the train-test resolution discrepancy
http://arxiv.org/abs/1906.06423
AUTHORS:        Hugo Touvron ; Andrea Vedaldi ; Matthijs Douze ; Hervé Jégou
HIGHLIGHT:      For instance, we obtain 77.1% top-1 accuracy on ImageNet with a ResNet-50 trained on 128x128 images, and 79.8% with one trained on 224x224 image.

35, TITLE:      Learning Canonical Shape Space for Category-Level 6D Object Pose and Size Estimation
http://arxiv.org/abs/2001.09322
AUTHORS:        Dengsheng Chen ; Jun Li ; Zheng Wang ; Kai Xu
HIGHLIGHT:      We present a novel approach to category-level 6D object pose and size estimation.

36, TITLE:      Fixing the train-test resolution discrepancy: FixEfficientNet
http://arxiv.org/abs/2003.08237
AUTHORS:        Hugo Touvron ; Andrea Vedaldi ; Matthijs Douze ; Hervé Jégou
HIGHLIGHT:      This note complements the paper "Fixing the train-test resolution discrepancy" that introduced the FixRes method.

37, TITLE:      Learning Texture Invariant Representation for Domain Adaptation of Semantic Segmentation
http://arxiv.org/abs/2003.00867
AUTHORS:        Myeongjin Kim ; Hyeran Byun
COMMENTS:       2020 CVPR
HIGHLIGHT:      In this paper, considering the fundamental difference between the two domains as the texture, we propose a method to adapt to the texture of the target domain.

38, TITLE:      Multi-Scale Progressive Fusion Network for Single Image Deraining
http://arxiv.org/abs/2003.10985
AUTHORS:        Kui Jiang ; Zhongyuan Wang ; Peng Yi ; Chen Chen ; Baojin Huang ; Yimin Luo ; Jiayi Ma ; Junjun Jiang
COMMENTS:       CVPR 2020
HIGHLIGHT:      In this work, we explore the multi-scale collaborative representation for rain streaks from the perspective of input image scales and hierarchical deep features in a unified framework, termed multi-scale progressive fusion network (MSPFN) for single image rain streak removal.

39, TITLE:      Towards Learning Structure via Consensus for Face Segmentation and Parsing
http://arxiv.org/abs/1911.00957
AUTHORS:        Iacopo Masi ; Joe Mathai ; Wael AbdAlmageed
COMMENTS:       To appear in the IEEE/CVF Conference on Computer Vision and Pattern Recognition CVPR 2020. Project page at https://github.com/isi-vista/structure_via_consensus
HIGHLIGHT:      Unlike state-of-the-art methods that treat face segmentation as an independent pixel prediction problem, we argue instead that it should hold highly correlated outputs within the same object pixels.

40, TITLE:      LoGAN: Latent Graph Co-Attention Network for Weakly-Supervised Video Moment Retrieval
http://arxiv.org/abs/1909.13784
AUTHORS:        Reuben Tan ; Huijuan Xu ; Kate Saenko ; Bryan A. Plummer
HIGHLIGHT:      To address this limitation, we propose a Latent Graph Co-Attention Network (LoGAN) that exploits fine-grained frame-by-word interactions to reason about correspondences between all possible pairs of frames, given the semantic context of the query.

41, TITLE:      3D Packing for Self-Supervised Monocular Depth Estimation
http://arxiv.org/abs/1905.02693
AUTHORS:        Vitor Guizilini ; Rares Ambrus ; Sudeep Pillai ; Allan Raventos ; Adrien Gaidon
HIGHLIGHT:      In this work, we propose a novel self-supervised monocular depth estimation method combining geometry with a new deep network, PackNet, learned only from unlabeled monocular videos.

42, TITLE:      SuperGlue: Learning Feature Matching with Graph Neural Networks
http://arxiv.org/abs/1911.11763
AUTHORS:        Paul-Edouard Sarlin ; Daniel DeTone ; Tomasz Malisiewicz ; Andrew Rabinovich
COMMENTS:       Oral at CVPR 2020, with appendix and link to publicly available code
HIGHLIGHT:      This paper introduces SuperGlue, a neural network that matches two sets of local features by jointly finding correspondences and rejecting non-matchable points.

43, TITLE:      TBT: Targeted Neural Network Attack with Bit Trojan
http://arxiv.org/abs/1909.05193
AUTHORS:        Adnan Siraj Rakin ; Zhezhi He ; Deliang Fan
HIGHLIGHT:      In this work, for the first time, we propose a novel Targeted Bit Trojan(TBT) method, which can insert a targeted neural Trojan into a DNN through the bit-flip attack.

44, TITLE:      Rapid Top-Down Synthesis of Large-Scale IoT Networks
http://arxiv.org/abs/2002.04244
AUTHORS:        Pradipta Ghosh ; Jonathan Bunton ; Dimitrios Pylorof ; Marcos Vieira ; Kevin Chan ; Ramesh Govindan ; Gaurav Sukhatme ; Paulo Tabuada ; Gunjan Verma
HIGHLIGHT:      Motivated by this, we consider the top-down synthesis of ad-hoc IoT networks for disaster response and search and rescue operations.

45, TITLE:      Deep Coarse-to-fine Dense Light Field Reconstruction with Flexible Sampling and Geometry-aware Fusion
http://arxiv.org/abs/1909.01341
AUTHORS:        Jing Jin ; Junhui Hou ; Jie Chen ; Huanqiang Zeng ; Sam Kwong ; Jingyi Yu
COMMENTS:       17 pages, 12 figures, 10 tables
HIGHLIGHT:      To this end, we propose a novel learning-based method, which accepts sparsely-sampled LFs with irregular structures, and produces densely-sampled LFs with arbitrary angular resolution accurately and efficiently.

46, TITLE:      Image Formation Model Guided Deep Image Super-Resolution
http://arxiv.org/abs/1908.06444
AUTHORS:        Jinshan Pan ; Yang Liu ; Deqing Sun ; Jimmy Ren ; Ming-Ming Cheng ; Jian Yang ; Jinhui Tang
COMMENTS:       AAAI 2020. The training code and models are available at https://github.com/jspan/PHYSICS SR
HIGHLIGHT:      We present a simple and effective image super-resolution algorithm that imposes an image formation constraint on the deep neural networks via pixel substitution.

47, TITLE:      Scalability in Perception for Autonomous Driving: Waymo Open Dataset
http://arxiv.org/abs/1912.04838
AUTHORS:        Pei Sun ; Henrik Kretzschmar ; Xerxes Dotiwalla ; Aurelien Chouard ; Vijaysai Patnaik ; Paul Tsui ; James Guo ; Yin Zhou ; Yuning Chai ; Benjamin Caine ; Vijay Vasudevan ; Wei Han ; Jiquan Ngiam ; Hang Zhao ; Aleksei Timofeev ; Scott Ettinger ; Maxim Krivokon ; Amy Gao ; Aditya Joshi ; Sheng Zhao ; Shuyang Cheng ; Yu Zhang ; Jonathon Shlens ; Zhifeng Chen ; Dragomir Anguelov
HIGHLIGHT:      In an effort to help align the research community's contributions with real-world self-driving problems, we introduce a new large scale, high quality, diverse dataset.

48, TITLE:      Cost Volume Pyramid Based Depth Inference for Multi-View Stereo
http://arxiv.org/abs/1912.08329
AUTHORS:        Jiayu Yang ; Wei Mao ; Jose M. Alvarez ; Miaomiao Liu
COMMENTS:       CVPR 2020 oral
HIGHLIGHT:      We propose a cost volume-based neural network for depth inference from multi-view images.

49, TITLE:      Robust Structured Group Local Sparse Tracker Using Deep Features
http://arxiv.org/abs/1902.07668
AUTHORS:        Mohammadreza Javanmardi ; Amir Hossein Farzaneh ; Xiaojun Qi
COMMENTS:       This submission is similar version of Structured Group Local Sparse Tracker arXiv:1902.06182
HIGHLIGHT:      In this paper, we propose a robust deep features-based structured group local sparse tracker (DF-SGLST), which exploits the deep features of local patches inside target candidates and represents them by a set of templates in the particle filter framework.

50, TITLE:      ASFD: Automatic and Scalable Face Detector
http://arxiv.org/abs/2003.11228
AUTHORS:        Bin Zhang ; Jian Li ; Yabiao Wang ; Ying Tai ; Chengjie Wang ; Jilin Li ; Feiyue Huang ; Yili Xia ; Wenjiang Pei ; Rongrong Ji
COMMENTS:       submit to ECCV2020
HIGHLIGHT:      In this paper, we propose a novel Automatic and Scalable Face Detector (ASFD), which is based on a combination of neural architecture search techniques as well as a new loss design.

51, TITLE:      CurlingNet: Compositional Learning between Images and Text for Fashion IQ Data
http://arxiv.org/abs/2003.12299
AUTHORS:        Youngjae Yu ; Seunghwan Lee ; Yuncheol Choi ; Gunhee Kim
COMMENTS:       4 pages, 4 figures, ICCV 2019 Linguistics Meets image and video retrieval workshop, Fashion IQ challenge
HIGHLIGHT:      We present an approach named CurlingNet that can measure the semantic distance of composition of image-text embedding.

52, TITLE:      BSP-Net: Generating Compact Meshes via Binary Space Partitioning
http://arxiv.org/abs/1911.06971
AUTHORS:        Zhiqin Chen ; Andrea Tagliasacchi ; Hao Zhang
COMMENTS:       Accepted to CVPR 2020 (Oral). Code is available at https://github.com/czq142857/BSP-NET-original
HIGHLIGHT:      Leading methods for learning generative models of shapes rely on implicit functions, and generate meshes only after expensive iso-surfacing routines.

53, TITLE:      DiverseDepth: Affine-invariant Depth Prediction Using Diverse Data
http://arxiv.org/abs/2002.00569
AUTHORS:        Wei Yin ; Xinlong Wang ; Chunhua Shen ; Yifan Liu ; Zhi Tian ; Songcen Xu ; Changming Sun ; Dou Renyin
COMMENTS:       Fixed typos
HIGHLIGHT:      We present a method for depth estimation with monocular images, which can predict high-quality depth on diverse scenes up to an affine transformation, thus preserving accurate shapes of a scene. In this work, we propose a dataset and methods to tackle this dilemma, aiming to predict accurate depth up to an affine transformation with good generalization to diverse scenes. First we construct a large-scale and diverse dataset, termed Diverse Scene Depth dataset (DiverseDepth), which has a broad range of scenes and foreground contents.

54, TITLE:      Deep learning achieves perfect anomaly detection on 108,308 retinal images including unlearned diseases
http://arxiv.org/abs/2001.05859
AUTHORS:        Ayaka Suzuki ; Yoshiro Suzuki
HIGHLIGHT:      As long as there is a risk of overlooking a disease with a technique, ophthalmologists must double-check even those images that the technique classifies as normal.

55, TITLE:      PU-GCN: Point Cloud Upsampling using Graph Convolutional Networks
http://arxiv.org/abs/1912.03264
AUTHORS:        Guocheng Qian ; Abdulellah Abualshour ; Guohao Li ; Ali Thabet ; Bernard Ghanem
COMMENTS:       The website and source code of this work are available at https://sites.google.com/kaust.edu.sa/pugcn and https://github.com/guochengqian/PU-GCN respectively
HIGHLIGHT:      We propose three novel point upsampling modules: Multi-branch GCN, Clone GCN, and NodeShuffle.

56, TITLE:      Multi-task Learning for Influence Estimation and Maximization
http://arxiv.org/abs/1904.08804
AUTHORS:        George Panagopoulos ; Fragkiskos D. Malliaros ; Michalis Vazirgiannis
COMMENTS:       11 pages
HIGHLIGHT:      We address the problem of influence maximization when the social network is accompanied by diffusion cascades.

57, TITLE:      Knowledge Graphs
http://arxiv.org/abs/2003.02320
AUTHORS:        Aidan Hogan ; Eva Blomqvist ; Michael Cochez ; Claudia d'Amato ; Gerard de Melo ; Claudio Gutierrez ; José Emilio Labra Gayo ; Sabrina Kirrane ; Sebastian Neumaier ; Axel Polleres ; Roberto Navigli ; Axel-Cyrille Ngonga Ngomo ; Sabbir M. Rashid ; Anisa Rula ; Lukas Schmelzeisen ; Juan Sequeda ; Steffen Staab ; Antoine Zimmermann
COMMENTS:       Revision from previous version: - Fixing flight companies in Figure 3 and changing some other details - Giving Figure 4 analogous data to Figure 3 for easier comparison - Updating discussion of the figures in Section 2.1.3. - Updating Example B.6 to reflect the new Figure 4. - Minor formatting change for Figure 27
HIGHLIGHT:      In this paper we provide a comprehensive introduction to knowledge graphs, which have recently garnered significant attention from both industry and academia in scenarios that require exploiting diverse, dynamic, large-scale collections of data.

58, TITLE:      Collective Embedding-based Entity Alignment via Adaptive Features
http://arxiv.org/abs/1912.08404
AUTHORS:        Wexin Zeng ; Xiang Zhao ; Jiuyang Tang ; Xuemin Lin
COMMENTS:       ICDE20
HIGHLIGHT:      To fill this gap, we propose a collective EA framework.

59, TITLE:      Knowledge Graph Alignment using String Edit Distance
http://arxiv.org/abs/2003.12145
AUTHORS:        Navdeep Kaur ; Gautam Kunapuli ; Sriraam Natarajan
COMMENTS:       Position Paper
HIGHLIGHT:      In this work, we propose a novel knowledge graph alignment technique based upon string edit distance that exploits the type information between entities and can find similarity between relations of any arity

60, TITLE:      Binary Classification of Alzheimer Disease using sMRI Imaging modality and Deep Learning
http://arxiv.org/abs/1809.06209
AUTHORS:        Ahsan Bin Tufail ; Qiu-Na Zhang ; Yong-Kui Ma
HIGHLIGHT:      Different from these methods, this paper proposes to construct multiple deep 2D convolutional neural networks (2D-CNNs) to learn the various features from local brain images which are combined to make the final classification for AD diagnosis.

61, TITLE:      Modeling the Background for Incremental Learning in Semantic Segmentation
http://arxiv.org/abs/2002.00718
AUTHORS:        Fabio Cermelli ; Massimiliano Mancini ; Samuel Rota Bulò ; Elisa Ricci ; Barbara Caputo
COMMENTS:       Accepted at CVPR 2020
HIGHLIGHT:      In this work we revisit classical incremental learning methods, proposing a new distillation-based framework which explicitly accounts for this shift.

62, TITLE:      Weakly-Supervised Action Localization by Generative Attention Modeling
http://arxiv.org/abs/2003.12424
AUTHORS:        Baifeng Shi ; Qi Dai ; Yadong Mu ; Jingdong Wang
COMMENTS:       CVPR2020. Code is available at https://github.com/bfshi/DGAM-Weakly-Supervised-Action-Localization
HIGHLIGHT:      To solve the problem, in this paper we propose to model the class-agnostic frame-wise probability conditioned on the frame attention using conditional Variational Auto-Encoder (VAE).

63, TITLE:      Unsupervised Sequence Forecasting of 100,000 Points for Unsupervised Trajectory Forecasting
http://arxiv.org/abs/2003.08376
AUTHORS:        Xinshuo Weng ; Jianren Wang ; Sergey Levine ; Kris Kitani ; Nicholas Rhinehart
COMMENTS:       Code will be available at https://github.com/xinshuoweng/SPCSF
HIGHLIGHT:      In this work, we study the problem of future prediction at the level of 3D scenes, represented by point clouds captured by a LiDAR sensor, i.e., directly learning to forecast the evolution of >100,000 points that comprise a complete scene.

64, TITLE:      Physics-Based Generative Adversarial Models for Image Restoration and Beyond
http://arxiv.org/abs/1808.00605
AUTHORS:        Jinshan Pan ; Jiangxin Dong ; Yang Liu ; Jiawei Zhang ; Jimmy Ren ; Jinhui Tang ; Yu-Wing Tai ; Ming-Hsuan Yang
COMMENTS:       IEEE TPAMI
HIGHLIGHT:      We present an algorithm to directly solve numerous image restoration problems (e.g., image deblurring, image dehazing, image deraining, etc.).

65, TITLE:      Hierarchically Robust Representation Learning
http://arxiv.org/abs/1911.04047
AUTHORS:        Qi Qian ; Juhua Hu ; Hao Li
COMMENTS:       accepted by CVPR'20
HIGHLIGHT:      In this work, we investigate this phenomenon and demonstrate that deep features can be suboptimal due to the fact that they are learned by minimizing the empirical risk.

66, TITLE:      EGO-TOPO: Environment Affordances from Egocentric Video
http://arxiv.org/abs/2001.04583
AUTHORS:        Tushar Nagarajan ; Yanghao Li ; Christoph Feichtenhofer ; Kristen Grauman
COMMENTS:       Published in CVPR 2020, project page: http://vision.cs.utexas.edu/projects/ego-topo/
HIGHLIGHT:      We introduce a model for environment affordances that is learned directly from egocentric video.

67, TITLE:      Algorithmic Guarantees for Inverse Imaging with Untrained Network Priors
http://arxiv.org/abs/1906.08763
AUTHORS:        Gauri Jagatap ; Chinmay Hegde
COMMENTS:       NeurIPS 2019 version with few modifications
HIGHLIGHT:      Specifically, we consider the problem of solving linear inverse problems, such as compressive sensing, as well as non-linear problems, such as compressive phase retrieval.

68, TITLE:      View-Invariant Probabilistic Embedding for Human Pose
http://arxiv.org/abs/1912.01001
AUTHORS:        Jennifer J. Sun ; Jiaping Zhao ; Liang-Chieh Chen ; Florian Schroff ; Hartwig Adam ; Ting Liu
HIGHLIGHT:      In this paper, we propose an approach for learning a compact view-invariant embedding space from 2D joint keypoints alone, without explicitly predicting 3D poses.

69, TITLE:      Insect pest image detection and recognition based on bio-inspired methods
http://arxiv.org/abs/1910.00296
AUTHORS:        Loris Nanni ; Gianluca Maguolo ; Fabio Pancino
HIGHLIGHT:      In this paper we propose an automatic classifier based on the fusion between saliency methods and convolutional neural networks.

70, TITLE:      Adversarial Domain Adaptation Being Aware of Class Relationships
http://arxiv.org/abs/1905.11931
AUTHORS:        Zeya Wang ; Baoyu Jing ; Yang Ni ; Nanqing Dong ; Pengtao Xie ; Eric P. Xing
HIGHLIGHT:      In this paper, we propose a novel relationship-aware adversarial domain adaptation (RADA) algorithm, which first utilizes a single multi-class domain discriminator to enforce the learning of inter-class dependency structure during domain-adversarial training and then aligns this structure with the inter-class dependencies that are characterized from training the label predictor on source domain.

71, TITLE:      Non-Adversarial Video Synthesis with Learned Priors
http://arxiv.org/abs/2003.09565
AUTHORS:        Abhishek Aich ; Akash Gupta ; Rameswar Panda ; Rakib Hyder ; M. Salman Asif ; Amit K. Roy-Chowdhury
COMMENTS:       Accepted to CVPR 2020
HIGHLIGHT:      Different from these methods, we focus on the problem of generating videos from latent noise vectors, without any reference input frames.

72, TITLE:      ParticleNet: Jet Tagging via Particle Clouds
http://arxiv.org/abs/1902.08570
AUTHORS:        Huilin Qu ; Loukas Gouskos
COMMENTS:       11 pages, 4 figures; v3: updated to match the version published in PRD; Code available at https://github.com/hqucms/ParticleNet
HIGHLIGHT:      Inspired by the notion of point clouds, we propose a new approach that considers a jet as an unordered set of its constituent particles, effectively a "particle cloud".

73, TITLE:      Stress Test Evaluation of Transformer-based Models in Natural Language Understanding Tasks
http://arxiv.org/abs/2002.06261
AUTHORS:        Carlos Aspillaga ; Andrés Carvallo ; Vladimir Araujo
COMMENTS:       Accepted paper LREC2020
HIGHLIGHT:      For that reason, this work evaluates three Transformer-based models (RoBERTa, XLNet, and BERT) in Natural Language Inference (NLI) and Question Answering (QA) tasks to know if they are more robust or if they have the same flaws as their predecessors.

74, TITLE:      AraBERT: Transformer-based Model for Arabic Language Understanding
http://arxiv.org/abs/2003.00104
AUTHORS:        Wissam Antoun ; Fady Baly ; Hazem Hajj
COMMENTS:       Proceedings of the Twelfth International Conference on Language Resources and Evaluation (LREC 2020), Marseille, France (2020)
HIGHLIGHT:      In this paper, we pre-trained BERT specifically for the Arabic language in the pursuit of achieving the same success that BERT did for the English language.

75, TITLE:      Attend to the beginning: A study on using bidirectional attention for extractive summarization
http://arxiv.org/abs/2002.03405
AUTHORS:        Ahmed Magooda ; Cezary Marcjan
COMMENTS:       To be published in FLAIRS33 (https://www.flairs-33.info/) and appear in he proceedings of AAAI
HIGHLIGHT:      In this work, we propose attending to the beginning of a document, to improve the performance of extractive summarization models when applied to forum discussion data.

76, TITLE:      A note on belief structures and S-approximation spaces
http://arxiv.org/abs/1805.10672
AUTHORS:        Ali Shakiba ; Amir Kafshdar Goharshady ; MohammadReza Hooshmandasl ; Mohsen Alambardar Meybodi
COMMENTS:       12 pages
HIGHLIGHT:      We study relations between evidence theory and S-approximation spaces.

77, TITLE:      Seeing without Looking: Contextual Rescoring of Object Detections for AP Maximization
http://arxiv.org/abs/1912.12290
AUTHORS:        Lourenço V. Pato ; Renato Negrinho ; Pedro M. Q. Aguiar
COMMENTS:       14 pages, 12 figures
HIGHLIGHT:      We propose to incorporate context in object detection by post-processing the output of an arbitrary detector to rescore the confidences of its detections.

78, TITLE:      Bundle Adjustment on a Graph Processor
http://arxiv.org/abs/2003.03134
AUTHORS:        Joseph Ortiz ; Mark Pupilli ; Stefan Leutenegger ; Andrew J. Davison
COMMENTS:       Published in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2020). Video: https://www.youtube.com/watch?v=TqeN8aQNgd0
HIGHLIGHT:      We show for the first time that the classical computer vision problem of bundle adjustment (BA) can be solved extremely fast on a graph processor using Gaussian Belief Propagation.

79, TITLE:      Assessing Image Quality Issues for Real-World Problems
http://arxiv.org/abs/2003.12511
AUTHORS:        Tai-Yin Chiu ; Yinan Zhao ; Danna Gurari
HIGHLIGHT:      We introduce a new large-scale dataset that links the assessment of image quality issues to two practical vision tasks: image captioning and visual question answering.

80, TITLE:      Explain Your Move: Understanding Agent Actions Using Salient and Relevant Feature Attribution
http://arxiv.org/abs/1912.12191
AUTHORS:        Nikaash Puri ; Sukriti Verma ; Piyush Gupta ; Dhruv Kayastha ; Shripad Deshmukh ; Balaji Krishnamurthy ; Sameer Singh
COMMENTS:       Accepted at the International Conference on Learning Representations (ICLR) 2020
HIGHLIGHT:      We compare SARFA with existing approaches on agents trained to play board games (Chess and Go) and Atari games (Breakout, Pong and Space Invaders).

81, TITLE:      Low-Budget Label Query through Domain Alignment Enforcement
http://arxiv.org/abs/2001.00238
AUTHORS:        Jurandy Almeida ; Cristiano Saltori ; Paolo Rota ; Nicu Sebe
HIGHLIGHT:      In this work, we tackle a new problem named low-budget label query that consists in suggesting to the user a small (low budget) set of samples to be labelled, from a completely unlabelled dataset, with the final goal of maximizing the classification accuracy on that dataset.

82, TITLE:      Towards Shape Biased Unsupervised Representation Learning for Domain Generalization
http://arxiv.org/abs/1909.08245
AUTHORS:        Nader Asadi ; Amir M. Sarfi ; Mehrdad Hosseinzadeh ; Zahra Karimpour ; Mahdi Eftekhari
COMMENTS:       Under review
HIGHLIGHT:      In this work, we propose a learning framework to improve the shape bias property of self-supervised methods.

83, TITLE:      The Liar's Walk: Detecting Deception with Gait and Gesture
http://arxiv.org/abs/1912.06874
AUTHORS:        Tanmay Randhavane ; Uttaran Bhattacharya ; Kyra Kapsaskis ; Kurt Gray ; Aniket Bera ; Dinesh Manocha
COMMENTS:       10 pages, 6 Figures
HIGHLIGHT:      We present a data-driven deep neural algorithm for detecting deceptive walking behavior using nonverbal cues like gaits and gestures. Additionally, we present a novel dataset, DeceptiveWalk, that contains gaits and gestures with their associated deception labels.

84, TITLE:      Towards Backward-Compatible Representation Learning
http://arxiv.org/abs/2003.11942
AUTHORS:        Yantao Shen ; Yuanjun Xiong ; Wei Xia ; Stefano Soatto
COMMENTS:       Accepted to CVPR 2020 as oral
HIGHLIGHT:      We propose a framework to train embedding models, called backward-compatible training (BCT), as a first step towards backward compatible representation learning.

85, TITLE:      BachGAN: High-Resolution Image Synthesis from Salient Object Layout
http://arxiv.org/abs/2003.11690
AUTHORS:        Yandong Li ; Yu Cheng ; Zhe Gan ; Licheng Yu ; Liqiang Wang ; Jingjing Liu
COMMENTS:       Accepted to CVPR 2020
HIGHLIGHT:      We propose a new task towards more practical application for image generation - high-quality image synthesis from salient object layout.

86, TITLE:      Gabor Layers Enhance Network Robustness
http://arxiv.org/abs/1912.05661
AUTHORS:        Juan C. Pérez ; Motasem Alfarra ; Guillaume Jeanneret ; Adel Bibi ; Ali Thabet ; Bernard Ghanem ; Pablo Arbeláez
COMMENTS:       32 pages, 23 figures, 14 tables
HIGHLIGHT:      We revisit the benefits of merging classical vision concepts with deep learning models.

87, TITLE:      Set-Constrained Viterbi for Set-Supervised Action Segmentation
http://arxiv.org/abs/2002.11925
AUTHORS:        Jun Li ; Sinisa Todorovic
COMMENTS:       CVPR 2020 conference paper
HIGHLIGHT:      Our first contribution is the formulation of a new set-constrained Viterbi algorithm (SCV).

88, TITLE:      Conv-MPN: Convolutional Message Passing Neural Network for Structured Outdoor Architecture Reconstruction
http://arxiv.org/abs/1912.01756
AUTHORS:        Fuyang Zhang ; Nelson Nauata ; Yasutaka Furukawa
COMMENTS:       Accepted by CVPR2020
HIGHLIGHT:      This paper proposes a novel message passing neural (MPN) architecture Conv-MPN, which reconstructs an outdoor building as a planar graph from a single RGB image.

89, TITLE:      Safety-Aware Hardening of 3D Object Detection Neural Network Systems
http://arxiv.org/abs/2003.11242
AUTHORS:        Chih-Hong Cheng
COMMENTS:       Withdraw work-in-progress research report. The use of KITTI dataset, under the intention of scientific publication and knowledge dissemination, shall still be clarified legally, whether it fits the definition of non-commercial use which is stated in Creative Commons
HIGHLIGHT:      Subsequently, we introduce a specialized loss function reflecting (1) the safety specification, (2) the use of single-stage detection architecture, and finally, (3) the characterization of robustness under perturbation.

90, TITLE:      What it Thinks is Important is Important: Robustness Transfers through Input Gradients
http://arxiv.org/abs/1912.05699
AUTHORS:        Alvin Chan ; Yi Tay ; Yew-Soon Ong
COMMENTS:       Accepted as Oral in CVPR 2020, Camera-Ready Version
HIGHLIGHT:      Through experiments in MNIST, CIFAR-10, CIFAR-100 and Tiny-ImageNet, we show that our proposed method, input gradient adversarial matching, can transfer robustness across different tasks and even across different model architectures.

91, TITLE:      BRPO: Batch Residual Policy Optimization
http://arxiv.org/abs/2002.05522
AUTHORS:        Sungryull Sohn ; Yinlam Chow ; Jayden Ooi ; Ofir Nachum ; Honglak Lee ; Ed Chi ; Craig Boutilier
HIGHLIGHT:      To remedy this, we propose residual policies, where the allowable deviation of the learned policy is state-action-dependent.

92, TITLE:      Cross-Domain Ambiguity Detection using Linear Transformation of Word Embedding Spaces
http://arxiv.org/abs/1910.12956
AUTHORS:        Vaibhav Jain ; Ruchika Malhotra ; Sanskar Jain ; Nishant Tanwar
COMMENTS:       Published as part of the proceedings of the 3rd Workshop on Natural Language Processing for Requirements Engineering in CEUR-WS Vol-2584
HIGHLIGHT:      This paper proposes a natural language processing approach to find potentially ambiguous words for a given set of domains.

93, TITLE:      Cross-lingual topic prediction for speech using translations
http://arxiv.org/abs/1908.11425
AUTHORS:        Sameer Bansal ; Herman Kamper ; Adam Lopez ; Sharon Goldwater
COMMENTS:       Accepted to ICASSP 2020
HIGHLIGHT:      Given a large amount of unannotated speech in a low-resource language, can we classify the speech utterances by topic?

94, TITLE:      Compound Probabilistic Context-Free Grammars for Grammar Induction
http://arxiv.org/abs/1906.10225
AUTHORS:        Yoon Kim ; Chris Dyer ; Alexander M. Rush
COMMENTS:       ACL 2019
HIGHLIGHT:      We study a formalization of the grammar induction problem that models sentences as being generated by a compound probabilistic context-free grammar.

95, TITLE:      AutoFCL: Automatically Tuning Fully Connected Layers for Handling Small Dataset
http://arxiv.org/abs/2001.11951
AUTHORS:        S. H. Shabbeer Basha ; Sravan Kumar Vinakota ; Shiv Ram Dubey ; Viswanath Pulabaigari ; Snehasis Mukherjee
HIGHLIGHT:      In this paper, the proposed AutoFCL model attempts to learn the structure of FC layers of a CNN automatically using Bayesian optimization.

96, TITLE:      Peeking into occluded joints: A novel framework for crowd pose estimation
http://arxiv.org/abs/2003.10506
AUTHORS:        Lingteng Qiu ; Xuanye Zhang ; Yanran Li ; Guanbin Li ; Xiaojun Wu ; Zixiang Xiong ; Xiaoguang Han ; Shuguang Cui
COMMENTS:       The code of OPEC-Net is available at: https://lingtengqiu.github.io/2020/03/22/OPEC-Net/
HIGHLIGHT:      Therefore, we thoroughly pursue this problem and propose a novel OPEC-Net framework together with a new Occluded Pose (OCPose) dataset with 9k annotated images.

97, TITLE:      Cross-modal Deep Face Normals with Deactivable Skip Connections
http://arxiv.org/abs/2003.09691
AUTHORS:        Victoria Fernandez Abrevaya ; Adnane Boukhayma ; Philip H. S. Torr ; Edmond Boyer
COMMENTS:       CVPR 2020
HIGHLIGHT:      We present an approach for estimating surface normals from in-the-wild color images of faces.

98, TITLE:      A Context-Aware Loss Function for Action Spotting in Soccer Videos
http://arxiv.org/abs/1912.01326
AUTHORS:        Anthony Cioppa ; Adrien Deliège ; Silvio Giancola ; Bernard Ghanem ; Marc Van Droogenbroeck ; Rikke Gade ; Thomas B. Moeslund
COMMENTS:       Accepted for CVPR2020 main conference. This document contains 8 pages + references + supplementary material
HIGHLIGHT:      In this paper, we propose a novel loss function that specifically considers the temporal context naturally present around each action, rather than focusing on the single annotated frame to spot.

99, TITLE:      Occlusion-robust Online Multi-object Visual Tracking using a GM-PHD Filter with CNN-based Re-identification
http://arxiv.org/abs/1912.05949
AUTHORS:        Nathanael L. Baisa
COMMENTS:       arXiv admin note: text overlap with arXiv:1908.03945
HIGHLIGHT:      We propose a novel online multi-object visual tracking algorithm via a tracking-by-detection paradigm using a Gaussian mixture Probability Hypothesis Density (GM-PHD) filter and deep Convolutional Neural Network (CNN) appearance representations learning.

100, TITLE:     D3VO: Deep Depth, Deep Pose and Deep Uncertainty for Monocular Visual Odometry
http://arxiv.org/abs/2003.01060
AUTHORS:        Nan Yang ; Lukas von Stumberg ; Rui Wang ; Daniel Cremers
COMMENTS:       CVPR 2020
HIGHLIGHT:      We propose D3VO as a novel framework for monocular visual odometry that exploits deep networks on three levels -- deep depth, pose and uncertainty estimation.

101, TITLE:     Domain adaptation for holistic skin detection
http://arxiv.org/abs/1903.06969
AUTHORS:        Aloisio Dourado ; Frederico Guth ; Teofilo Emidio de Campos ; Li Weigang
COMMENTS:       11 pages, 10 figures, 6 tables
HIGHLIGHT:      In this paper we present a comprehensive evaluation of holistic and local Convolutional Neural Network (CNN) approaches on in-domain and cross-domain experiments and compare with state-of-the-art pixel-based approaches.

102, TITLE:     Distilling Knowledge from Graph Convolutional Networks
http://arxiv.org/abs/2003.10477
AUTHORS:        Yiding Yang ; Jiayan Qiu ; Mingli Song ; Dacheng Tao ; Xinchao Wang
COMMENTS:       Accepted by CVPR 2020
HIGHLIGHT:      In this paper, we propose to our best knowledge the first dedicated approach to {distilling} knowledge from a pre-trained GCN model.

103, TITLE:     ViBE: Dressing for Diverse Body Shapes
http://arxiv.org/abs/1912.06697
AUTHORS:        Wei-Lin Hsiao ; Kristen Grauman
COMMENTS:       Accepted to CVPR 2020
HIGHLIGHT:      We introduce ViBE, a VIsual Body-aware Embedding that captures clothing's affinity with different body shapes.

104, TITLE:     DR Loss: Improving Object Detection by Distributional Ranking
http://arxiv.org/abs/1907.10156
AUTHORS:        Qi Qian ; Lei Chen ; Hao Li ; Rong Jin
COMMENTS:       accepted by CVPR'20
HIGHLIGHT:      In this work, we propose a novel distributional ranking (DR) loss to handle the challenge.

105, TITLE:     DR-KFS: A Differentiable Visual Similarity Metric for 3D Shape Reconstruction
http://arxiv.org/abs/1911.09204
AUTHORS:        Jiongchao Jin ; Akshay Gadi Patil ; Zhang Xiong ; Hao Zhang
HIGHLIGHT:      We introduce a differential visual similarity metric to train deep neural networks for 3D reconstruction, aimed at improving reconstruction quality.

106, TITLE:     $ε$-shotgun: $ε$-greedy Batch Bayesian Optimisation
http://arxiv.org/abs/2002.01873
AUTHORS:        George De Ath ; Richard M. Everson ; Jonathan E. Fieldsend ; Alma A. M. Rahat
COMMENTS:       Genetic and Evolutionary Computation Conference 2020 (GECCO '20). 9 pages (main paper) + 11 pages (supplementary material). Code avaliable at https://github.com/georgedeath/eshotgun
HIGHLIGHT:      We present an $\epsilon$-greedy procedure for Bayesian optimisation in batch settings in which the black-box function can be evaluated multiple times in parallel.

107, TITLE:     Generalizing from a Few Examples: A Survey on Few-Shot Learning
http://arxiv.org/abs/1904.05046
AUTHORS:        Yaqing Wang ; Quanming Yao ; James Kwok ; Lionel M. Ni
HIGHLIGHT:      In this paper, we conduct a thorough survey to fully understand FSL.

108, TITLE:     Rethinking Softmax with Cross-Entropy: Neural Network Classifier as Mutual Information Estimator
http://arxiv.org/abs/1911.10688
AUTHORS:        Zhenyue Qin ; Dongwoo Kim ; Tom Gedeon
HIGHLIGHT:      To this end, we propose infoCAM, informative class activation map, which highlights regions of the input image that are the most relevant to a given label based on differences in information.

109, TITLE:     Quantum algorithms and approximating polynomials for composed functions with shared inputs
http://arxiv.org/abs/1809.02254
AUTHORS:        Mark Bun ; Robin Kothari ; Justin Thaler
COMMENTS:       31 pages; 1 figure. This update includes an additional result on lower bounds for AC$^0 \circ \oplus$ computing the Inner Product function on average
HIGHLIGHT:      We give new quantum algorithms for evaluating composed functions whose inputs may be shared between bottom-level gates.

110, TITLE:     See and Read: Detecting Depression Symptoms in Higher Education Students Using Multimodal Social Media Data
http://arxiv.org/abs/1912.01131
AUTHORS:        Paulo Mann ; Aline Paes ; Elton H. Matsushima
COMMENTS:       This article was accepted (15 November 2019) and will appear in the proceedings of ICWSM 2020
HIGHLIGHT:      In this work, we focus on detecting the severity of the depression symptoms in higher education students, by comparing deep learning to feature engineering models induced from both the pictures and their captions posted on Instagram.

111, TITLE:     Comprehensive Named Entity Recognition on CORD-19 with Distant or Weak Supervision
http://arxiv.org/abs/2003.12218
AUTHORS:        Xuan Wang ; Xiangchen Song ; Yingjun Guan ; Bangzheng Li ; Jiawei Han
HIGHLIGHT:      We also hope this dataset can bring insights for the COVID- 19 studies, both on the biomedical side and on the social side. We created this CORD-19-NER dataset with comprehensive named entity recognition (NER) on the COVID-19 Open Research Dataset Challenge (CORD-19) corpus (2020- 03-13).

112, TITLE:     Learning Multi-Object Tracking and Segmentation from Automatic Annotations
http://arxiv.org/abs/1912.02096
AUTHORS:        Lorenzo Porzi ; Markus Hofinger ; Idoia Ruiz ; Joan Serrat ; Samuel Rota Bulò ; Peter Kontschieder
HIGHLIGHT:      In this work we contribute a novel pipeline to automatically generate training data, and to improve over state-of-the-art multi-object tracking and segmentation (MOTS) methods.

113, TITLE:     Encoder-Decoder Based Convolutional Neural Networks with Multi-Scale-Aware Modules for Crowd Counting
http://arxiv.org/abs/2003.05586
AUTHORS:        Pongpisit Thanasutives ; Ken-ichi Fukui ; Masayuki Numao ; Boonserm Kijsirikul
COMMENTS:       V3: Change to the IEEE format
HIGHLIGHT:      In this paper, we proposed two modified neural network architectures based on SFANet and SegNet respectively for accurate and efficient crowd counting.

114, TITLE:     Reusing Discriminators for Encoding: Towards Unsupervised Image-to-Image Translation
http://arxiv.org/abs/2003.00273
AUTHORS:        Runfa Chen ; Wenbing Huang ; Binghui Huang ; Fuchun Sun ; Bin Fang
COMMENTS:       Accepted to CVPR 2020
HIGHLIGHT:      To tackle this issue, we develop a decoupled training strategy by which the encoder is only trained when maximizing the adversary loss while keeping frozen otherwise.

115, TITLE:     PADS: Policy-Adapted Sampling for Visual Similarity Learning
http://arxiv.org/abs/2003.11113
AUTHORS:        Karsten Roth ; Timo Milbich ; Björn Ommer
COMMENTS:       Accepted to CVPR2020
HIGHLIGHT:      We, therefore, employ reinforcement learning and have a teacher network adjust the sampling distribution based on the current state of the learner network, which represents visual similarity.

116, TITLE:     DADA: Differentiable Automatic Data Augmentation
http://arxiv.org/abs/2003.03780
AUTHORS:        Yonggang Li ; Guosheng Hu ; Yongtao Wang ; Timothy Hospedales ; Neil M. Robertson ; Yongxin Yang
HIGHLIGHT:      In this paper, we propose Differentiable Automatic Data Augmentation (DADA) which dramatically reduces the cost.

117, TITLE:     The Garden of Forking Paths: Towards Multi-Future Trajectory Prediction
http://arxiv.org/abs/1912.06445
AUTHORS:        Junwei Liang ; Lu Jiang ; Kevin Murphy ; Ting Yu ; Alexander Hauptmann
COMMENTS:       CVPR 2020. Code, models and dataset are available at: https://next.cs.cmu.edu/multiverse/index.html
HIGHLIGHT:      This paper studies the problem of predicting the distribution over multiple possible future paths of people as they move through various visual scenes.

118, TITLE:     Distilling Effective Supervision from Severe Label Noise
http://arxiv.org/abs/1910.00701
AUTHORS:        Zizhao Zhang ; Han Zhang ; Sercan O. Arik ; Honglak Lee ; Tomas Pfister
HIGHLIGHT:      We present a holistic framework to train deep neural networks in a way that is highly invulnerable to label noise.

119, TITLE:     Histogram Layers for Texture Analysis
http://arxiv.org/abs/2001.00215
AUTHORS:        Joshua Peeples ; Weihuang Xu ; Alina Zare
COMMENTS:       16 pages, 6 figures
HIGHLIGHT:      We present a histogram layer for artificial neural networks (ANNs).

120, TITLE:     Few-Shot Learning via Embedding Adaptation with Set-to-Set Functions
http://arxiv.org/abs/1812.03664
AUTHORS:        Han-Jia Ye ; Hexiang Hu ; De-Chuan Zhan ; Fei Sha
COMMENTS:       Accepted by CVPR 2020; The code is available at https://github.com/Sha-Lab/FEAT
HIGHLIGHT:      In this paper, we propose a novel approach to adapt the instance embeddings to the target classification task with a set-to-set function, yielding embeddings that are task-specific and are discriminative.

121, TITLE:     DIDFuse: Deep Image Decomposition for Infrared and Visible Image Fusion
http://arxiv.org/abs/2003.09210
AUTHORS:        Zixiang Zhao ; Shuang Xu ; Chunxia Zhang ; Junmin Liu ; Pengfei Li ; Jiangshe Zhang
COMMENTS:       We hope to release it later
HIGHLIGHT:      This paper proposes a novel auto-encoder (AE) based fusion network.