Pytorch implementation of HighRes-net, a neural network for multi-frame super-resolution, trained and tested on the European Space Agency’s Kelvin competition. This is a ServiceNow Research project that was started at Element AI.

Pytorch code for "BSRT: Improving Burst Super-Resolution with Swin Transformer and Flow-Guided Deformable Alignment", CVPRW, 1st place in NTIRE 2022 BurstSR Challenge (real-world track).

Track oriented, multi target, multi hypothesis tracker

Unsupervised Learning of Multi-Frame Optical Flow with Occlusions

✨ [CVPR 2023] NeuralPCI: Spatio-temporal Neural Field for 3D Point Cloud Multi-frame Non-linear Interpolation

deep learning models trained to denoise/deblur text images (signle frame, multi-frame) [pytorch]

SAFNet: Selective Alignment Fusion Network for Efficient HDR Imaging (ECCV 2024)

Code for "Spatial-Temporal Enhanced Transformer Towards Multi-Frame 3D Object Detection"

MT4 proxy terminal can be used for protection of Expert, Indicators and Scripts as well as for multi-timeframe testing and signals distribution

App developed in collaboration with a Neuroscientist, specifically engineered for image compression with emphasis on maintaining high image quality. It is tailored to ensure that the integrity of the images is preserved during the compression process.