10+ Codebases with 100+ Algorithms and 600+ Models !

Open Source Projects

Open source projects for academic research of computer vision. Support a wide range of research topics, e.g. detection, classification, super resolution.

Highlighted Projects

MMCV

A foundational python library for computer vision
Learn more

MMDetection

Object detection toolbox and benchmark
Learn more

MMAction2

Next generation toolbox for action understanding
Learn more

MMEditing

Image and video editing toolbox for editing tasks
Learn more

MMDetection3D

The next generation toolbox for general 3D detection
Learn more

MMSegmentation

Comprehensive semantic segmentation toolbox
Learn more
20+ Datasets with Benchmarks !

Open Datasets

In OpenMMLab, we also provide 20+ datasets owned by MMLab. The datasets with annotations are now easy to access at OpenMMLab.

Highlighted Datasets

FineGym

Dataset for fine-grained action understanding
Learn more

DeepFashion

Dataset on fashion analysis and synthesis
Learn more

MovieNet

A holistic dataset for movie understanding
Learn more

DeeperForensics

Dataset for real-world face forgery detection
Learn more

Placepedia

Dataset for place, landmark and area analysis
Learn more

WIDER

Faces and bodies detection, identities recognition
Learn more

Documentations

Comprehensive documentation includes codebase instructions, dataset usages and tutorials for new beginners.

9
Codebase / Dataset Documentations

We've wrote documentations for 9 popular OpenMMLab codebases/datasets for users to easily start with the codes/datasets.

9
Easy-to-follow Tutorials

Tutorials are also provided for new beginners to quick start with various research topics in computer vision field.

100+
Classical / Latest Algorithms

Within the documentations and tutorials, about 100+ either classic or lateset algorithms are introduced in detail.

View Documentations

Organizational Contributors