The project for NTU's course on Machine Learning, CZ4041

Comparison between different DL models such as VGGnet,InceptionV3,Resnet for copy move forgery detection

Detect Deepfaked Faces Using Multiple Deeplearning Models

Using an External dataset to get the pre-trained weights of the NIH dataset and training on the provided dataset to detect the presence of pneumonia.

The purpose of this project is to develop an AI-powered system capable of detecting deepfake facial data in biometric systems. By leveraging machine learning, specifically XceptionNet architecture, the project aims to classify facial data as real or fake with high accuracy and reliability.

Improved Deep Learning Model has been used to classify Breast Cancer from Histopathological Tissue Images.

Implementation of some basic Image Annotation methods (using various loss functions & threshold optimization) on Corel-5k dataset with PyTorch library

Flower image classification using Transfer learning (Xception)

IPython Notebook to build the model for Dog Breed Prediction

The project focuses on classifying brain tumors using the Multi-Modal Squeeze and Excitation Network.

Development and analysis of various deep NN models to detect glaucoma cases from fundus images. The performance of the best model was evaluated with cross-validation. Mean F1-score: 0.95975, with a standard deviation of 0.02274.

In this project, we used a transfer learning approach to build an image classification model for the classification of skin lesion, we trained our model specifically on the ham10000 dataset available on kaggle and we were able to achieve a 93.6% accuracy

This repository hosts the Cervical Cancer Image Classification project, a comprehensive effort aimed at improving the classification accuracy of Squamous Cell Carcinoma (SCC) through advanced deep learning models and ensemble techniques. The project utilizes the Herlev dataset.