• Stars
    star
    1,847
  • Rank 24,199 (Top 0.5 %)
  • Language
    Python
  • License
    Apache License 2.0
  • Created almost 5 years ago
  • Updated about 2 years ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

An open-source toolbox for action understanding based on PyTorch

MMAction

Introduction

MMAction is an open source toolbox for action understanding based on PyTorch. It is a part of the open-mmlab project developed by Multimedia Laboratory, CUHK.

Major Features

  • MMAction is capable of dealing with all of the tasks below.

    • action recognition from trimmed videos
    • temporal action detection (also known as action localization) in untrimmed videos
    • spatial-temporal action detection in untrimmed videos.
  • Support for various datasets

    Video datasets have emerging throughout the recent years and have greatly fostered the devlopment of this field. MMAction provides tools to deal with various datasets.

  • Support for multiple action understanding frameworks

    MMAction implements popular frameworks for action understanding:

    • For action recognition, various algorithms are implemented, including TSN, I3D, SlowFast, R(2+1)D, CSN.
    • For temporal action detection, we implement SSN.
    • For spatial temporal atomic action detection, a Fast-RCNN baseline is provided.
  • Modular design

    The tasks in human action understanding share some common aspects such as backbones, and long-term and short-term sampling schemes. Also, tasks can benefit from each other. For example, a better backbone for action recognition will bring performance gain for action detection. Modular design enables us to view action understanding in a more integrated perspective.

License

The project is release under the Apache 2.0 license.

Updates

OmniSource Model Release (22/08/2020)

  • We release several models of our work OmniSource. These models are jointly trained with Kinetics-400 and OmniSourced web dataset. Those models are of good performance (Top1 Accuracy: 75.7% for 3-segment TSN and 80.4% for SlowOnly on Kinetics-400 val) and the learned representation transfer well to other tasks.

v0.2.0 (15/03/2020)

  • We build a diversified modelzoo for action recognition, which include popular algorithms (TSN, I3D, SlowFast, R(2+1)D, CSN). The performance is aligned with or better than the original papers.

v0.1.0 (19/06/2019)

  • MMAction is online!

Model zoo

Results and reference models are available in the model zoo.

Installation

Please refer to INSTALL.md for installation.

Update: for Docker installation, Please refer to DOCKER.md for using docker for this project.

Data preparation

Please refer to DATASET.md for a general knowledge of data preparation. Detailed documents for the supported datasets are available in data_tools/.

Get started

Please refer to GETTING_STARTED.md for detailed examples and abstract usage.

Contributing

We appreciate all contributions to improve MMAction. Please refer to CONTRUBUTING.md for the contributing guideline.

Citation

If you use our codebase or models in your research, please cite this work. We will release a technical report later.

@misc{mmaction2019,
  author =       {Yue Zhao, Yuanjun Xiong, Dahua Lin},
  title =        {MMAction},
  howpublished = {\url{https://github.com/open-mmlab/mmaction}},
  year =         {2019}
}

Contact

If you have any question, please file an issue or contact the author:

More Repositories

1

mmdetection

OpenMMLab Detection Toolbox and Benchmark
Python
27,899
star
2

mmsegmentation

OpenMMLab Semantic Segmentation Toolbox and Benchmark.
Python
7,372
star
3

mmagic

OpenMMLab Multimodal Advanced, Generative, and Intelligent Creation Toolbox. Unlock the magic 🪄: Generative-AI (AIGC), easy-to-use APIs, awsome model zoo, diffusion models, for text-to-image generation, image/video restoration/enhancement, etc.
Jupyter Notebook
6,575
star
4

mmcv

OpenMMLab Computer Vision Foundation
Python
5,626
star
5

mmpose

OpenMMLab Pose Estimation Toolbox and Benchmark.
Python
4,969
star
6

mmdetection3d

OpenMMLab's next-generation platform for general 3D object detection.
Python
4,781
star
7

OpenPCDet

OpenPCDet Toolbox for LiDAR-based 3D Object Detection.
Python
4,342
star
8

mmocr

OpenMMLab Text Detection, Recognition and Understanding Toolbox
Python
4,099
star
9

Amphion

Amphion (/æmˈfaɪən/) is a toolkit for Audio, Music, and Speech Generation. Its purpose is to support reproducible research and help junior researchers and engineers get started in the field of audio, music, and speech generation research and development.
Python
3,947
star
10

mmaction2

OpenMMLab's Next Generation Video Understanding Toolbox and Benchmark
Python
3,903
star
11

mmtracking

OpenMMLab Video Perception Toolbox. It supports Video Object Detection (VID), Multiple Object Tracking (MOT), Single Object Tracking (SOT), Video Instance Segmentation (VIS) with a unified framework.
Python
3,384
star
12

mmpretrain

OpenMMLab Pre-training Toolbox and Benchmark
Python
3,138
star
13

mmselfsup

OpenMMLab Self-Supervised Learning Toolbox and Benchmark
Python
3,084
star
14

mmskeleton

A OpenMMLAB toolbox for human pose estimation, skeleton-based action recognition, and action synthesis.
Python
2,855
star
15

mmyolo

OpenMMLab YOLO series toolbox and benchmark. Implemented RTMDet, RTMDet-Rotated,YOLOv5, YOLOv6, YOLOv7, YOLOv8,YOLOX, PPYOLOE, etc.
Python
2,719
star
16

mmdeploy

OpenMMLab Model Deployment Framework
Python
2,520
star
17

mmgeneration

MMGeneration is a powerful toolkit for generative models, based on PyTorch and MMCV.
Python
1,796
star
18

mmrotate

OpenMMLab Rotated Object Detection Toolbox and Benchmark
Python
1,724
star
19

Multimodal-GPT

Multimodal-GPT
Python
1,397
star
20

mmrazor

OpenMMLab Model Compression Toolbox and Benchmark.
Python
1,366
star
21

mmfashion

Open-source toolbox for visual fashion analysis based on PyTorch
Python
1,204
star
22

mmhuman3d

OpenMMLab 3D Human Parametric Model Toolbox and Benchmark
Python
1,129
star
23

mmengine

OpenMMLab Foundational Library for Training Deep Learning Models
Python
1,029
star
24

playground

A central hub for gathering and showcasing amazing projects that extend OpenMMLab with SAM and other exciting features.
Python
1,017
star
25

OpenMMLabCourse

OpenMMLab course index and stuff
Jupyter Notebook
919
star
26

mmflow

OpenMMLab optical flow toolbox and benchmark
Python
889
star
27

PIA

[CVPR 2024] PIA, your Personalized Image Animator. Animate your images by text prompt, combing with Dreambooth, achieving stunning videos. PIA,你的个性化图像动画生成器,利用文本提示将图像变为奇妙的动画
Python
716
star
28

mmfewshot

OpenMMLab FewShot Learning Toolbox and Benchmark
Python
662
star
29

awesome-vit

396
star
30

OpenUnReID

PyTorch open-source toolbox for unsupervised or domain adaptive object re-ID.
Python
389
star
31

labelbee-client

Out-of-the-box Annotation Toolbox
JavaScript
374
star
32

mim

MIM Installs OpenMMLab Packages
Python
329
star
33

denseflow

Extracting optical flow and frames
C++
286
star
34

MMGEN-FaceStylor

Python
245
star
35

mmeval

A unified evaluation library for multiple machine learning libraries
Python
243
star
36

labelbee

LabelBee is an annotation Library
TypeScript
223
star
37

OpenMMLabCamp

Jupyter Notebook
92
star
38

polynet

The Github Repo for PolyNet
77
star
39

CLUE

C++ Lightweight Utility Extensions
C++
70
star
40

mim-example

Python
57
star
41

mmengine-template

Python
47
star
42

ecosystem

36
star
43

mmstyles

Latex style file to facilitate writing of technical papers
TeX
36
star
44

mmpose-webcam-demo

Python
23
star
45

pre-commit-hooks

Python
17
star
46

mdformat-openmmlab

Python
8
star
47

.github

3
star