Deep Learning Computer Vision Algorithms for Real-World Use

PyTorch Blog Post On Image Similarity Search

DocEnTr: An end-to-end document image enhancement transformer - ICPR 2022

Enhanced protein mutational sampling using time-lagged variational autoencoders

Master thesis: Structured Auto-Encoder with application to Music Genre Recognition (code)

This is the code of experiments in paper Cross-Domain Adversarial Auto-Encoder(https://arxiv.org/abs/1804.06078)

Used the Functional API to built custom layers and non-sequential model types in TensorFlow, performed object detection, image segmentation, and interpretation of convolutions. Used generative deep learning including Auto Encoding, VAEs, and GANs to create new content.

Auto encoders based recommendation system

Point Cloud Gaussian Mixture Models and Shape Completion with Auto-encoder embeddings

Master thesis: Structured Auto-Encoder with application to Music Genre Recognition (report)

Modified and example codes of GAN in pytorch

My research work with a proof of concept for Image Restoration of motion-blurred images in Real-time using data augmentation and specific architecture of Deep Autoencoder network (inspired from U-Net model) with CNN layers. (Studied extensive use of functional APIs for custom layers, loss, and metrics, effects of regularization & Hyperparams opt…

simple VAE pytorch implementation

Master thesis: Structured Auto-Encoder with application to Music Genre Recognition (results)

Explore Network Anomaly Detection Project 📊💻. It achieves an exceptional 99.7% accuracy through a blend of supervised and unsupervised learning, extensive feature selection, and model experimentation. Stunning data visualizations using synthetic network traffic data offer insightful representations of anomalies, enhancing network security.

In this repository, denoising of medical images has been performed using convolutional auto-encoder models.