Found 106 repositories(showing 30)
open-mmlab
OpenMMLab's Next Generation Video Understanding Toolbox and Benchmark
isLinXu
💡💡💡awesome compute vision app in gradio
Wenhai-Zhu
No description available
kaan0821
Introduced 3D CBAM Attention to the PoseC3D in MMAction2
Whiffe
No description available
PINTO0309
ONNX export environment for mmaction2
Shu-Ang
自定义AVA数据集并使用mmaction2框架训练slowfast模型 copy from Whiffe
mohdshadaab
A Human Action Recognition pipeline using MMAction2 and kinetics400 dataset. MMAction2 is an open-source toolbox for video understanding based on PyTorch.
lyooyl
Make AVADataset custom dataset.
UoA-CARES
Researching sign language recognition using MMAction2
spped2000
for visualize joints and limbs heatmap to mmaction2
Hank-coder
No description available
hienhayho
Human Action Recogition With MMAction2
aidanboyne
Seizure detection using MMAction2 Framework
irvingzhang0512
No description available
higotenda
A framework for multi-modal media summarization.
No description available
namburusiddhartha
This repo builds on base I3D model provided by mmaction2, recognizing and localizing an activity in a video.
ntduong31
A production-ready behavior analysis pipeline integrating YOLOv8 for human detection and SOTA MMAction2 models to recognize over 700 human actions with rich visualization.
ahaqu01
基于mmaction2，重构了基于视频段和待检测人bboxs的动作识别算法接口
PINTO0309
Execution environment of mmaction2
thalangi33
Motion-guided sampler for Action Recognition using mmaction2
sherylchenn
mmaction2 (3D vision, action detection) with Auburn University Data Science Lab
Whiffe
No description available
No description available
No description available
shshjhjh4455
No description available
FIAS: A comprehensive framework for explainable, real-time fitness coaching. Features mmaction2 scripts, Grad-CAM visualizations, and an LLM-powered demo. An NTHU CVLAB graduation project.
maedmatt
This Python script uses MMAction2 to perform action recognition on video files. It can classify human actions in videos using a pre-trained model from the Kinetics-400 dataset.
jjoseph-ml
A comprehensive machine learning pipeline for seizure detection and classification using pose-based video analysis. This project implements a complete k-fold cross-validation workflow using MMAction2 framework with STGCN (Spatial-Temporal Graph Convolutional Networks) for temporal action recognition.