#310 · Primary category: Computer Vision
VIBE
Official implementation of CVPR2020 paper "VIBE: Video Inference for Human Body Pose and Shape Estimation"
Project last updated:03/24/23
GitHub Stars
3.2K
Forks
579
Contributors
8
License
Other
Why we included this project
VIBE is the official implementation of the CVPR 2020 paper of the same name, and it remains a solid reference point if you need 3D body pose and shape from ordinary video. The method walks through each frame of a clip and returns SMPL body-model parameters, so you get a full articulated body rather than 2D keypoints, which is what animation, biomechanics, sports tracking, and AR/VR work usually need. Its main practical edge over single-frame estimators is temporal: a recurrent structure carries information across frames, and that visibly reduces the jitter that shows up in real footage. The repository includes both demo and training code in PyTorch plus a Colab notebook, so you can try it on your own clips without assembling a pipeline from scratch. One caveat before you commit: the code and pretrained models ship under a non-commercial research license and depend on the SMPL body model, which has its own terms.
Articles for this project
No articles for this project yet.
To suggest a topic or contribute an article, contact us.
Related projects in this category
opencv
Open Source Computer Vision Library
RuView
π RuView turns commodity WiFi signals into real-time spatial intelligence, vital sign monitoring, and presence detection — all without a single pixel of video.
PaddleOCR
Turn any PDF or image document into structured data for your AI. A powerful, lightweight OCR toolkit that bridges the gap between images/PDFs and LLMs. Supports 100+ languages.
MinerU
Transforms complex documents like PDFs and Office docs into LLM-ready markdown/JSON for your Agentic workflows.
tesseract
Tesseract Open Source OCR Engine (main repository)