Human action classification system with pose-based (MediaPipe) and video-based (3D CNN) models. Features 100+ architectures for real-time pose classification and temporal models pretrained on UCF-101/HMDB51. Applications include autonomous vehicles, video surveillance, and action recognition research.

Two-stream CNNs for video action recognition implemented in Keras

My re-implementation of two stream action recognition

Video Recognition using Mixed Convolutional Tube (MiCT) on PyTorch with a ResNet backbone

This repository contains the code implementation used in the paper Temporally Coherent Embeddings for Self-Supervised Video Representation Learning (TCE).

Transformer for Action Recognition in PyTorch

Contains additional materials for two keras.io blog posts.

PyTorch Implementation of Attention Prompt Tuning: Parameter-Efficient Adaptation of Pre-Trained Models for Action Recognition

This repository contains my personal code for the paper Learning Spatiotemporal Features with 3D Convolutional Networks by Du Tran, Lubomir Bourdev, Rob Fergus, Lorenzo Torresani, Manohar Paluri.

Simple Action Recognition experimentation with the UCF101 Dataset and EfficientNets.

This is the project repository for the research study "Skeleton-Split Framework using Spatial Temporal Graph Convolutional Networks for Action Recognition" presented by Motasem S. Alsawadi and Miguel Rio.

This repository dedicated to Adaptive Deep Learning for Environment-Agnostic Human Action Recognition. This project focuses on developing a robust deep learning system tailored for accurate identification and analysis of human actions across diverse environments, with applications spanning surveillance, security, sports, and fitness.

Salient Object Detection for Video Masked Auto-Encoders

This project builds a video classification model using CNNs for spatial feature extraction and RNNs for temporal sequence modeling. Utilizing the UCF101 dataset, it covers data preprocessing, feature extraction, model training, and evaluation, providing a comprehensive approach to action recognition in videos.

Project work of Cognitive Computing and Artificial Intelligence course of University of Catania.

Implementation of I3D in PyTorch altered for EDR experiments

Classify UCF101 videos using one frame at a time with a CNN(InceptionV3)