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.

Video Platform for Action Recognition and Object Detection in Pytorch

Implementation Code of the paper Optical Flow Guided Feature, CVPR 2018

Caffe implementation for "Hidden Two-Stream Convolutional Networks for Action Recognition"

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

Video Representation Learning by Recognizing Temporal Transformations. In ECCV, 2020.

Use 3D ResNet to extract features of UCF101 and HMDB51 and then classify them.

[AAAI 2023 (Oral)] CrissCross: Self-Supervised Audio-Visual Representation Learning with Relaxed Cross-Modal Synchronicity

Actor-agnostic Multi-label Action Recognition with Multi-modal Query [ICCVW '23]

由视频数据集生成optical flow，frames，warped flow

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

[AAAI 2024] XKD: Cross-modal Knowledge Distillation with Domain Alignment for Video Representation Learning.

Two-stream CNNs for Video Action Recognition using Stacked Optical Flow. Implemented in Keras on HMDB-51 dataset.

Temporal 3D ConvNet

Salient Video Frames Sampling Method Using the Mean of Deep Features for Efficient Model Training (KIBME 2021)

Human Activity Recognition (HAR) using LRCN approach and MoveNet approach on HMDB-51 dataset

Action Recognition using a two stream CNN architecture with Frames and Optical Flows. Realized using Keras on HMDB51 dataset.

A project on video classification using PyTorch 2

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.

Action Recognition Using CNN + Bidirectional RNN