I3D implemetation in Keras + video preprocessing + visualization of results

A Pytorch implementation of The Visual Centrifuge: Model-Free Layered Video Representations.

Detection of temporal boundaries in sign language videos, as part of the Object Recognition & Computer Vision course in the MVA master program.

GRU Model That uses a sequence of MobileNet Image Features to classify a video clip as class label shoplifting or not shoplifting.

Inception 3D in keras for Indian Sign Language.

This application is designed to recognize and translate sign language gestures into text. It utilizes four pre-trained models, each containing different vocabulary sizes (100, 300, 1000, and 2000 words)

Convert videos into IADs