Top Repositories
Markerless kinematics with any cameras — From 2D Pose estimation to 3D OpenSim motion
Source code of the marker enhancer
Momentum Human Rig is an anatomically-inspired parametric full-body digital human model developed at Meta. It includes: A parametric body skeletal model; A realistic 3D mesh skinned to the skeleton with levels of detail;A body blendshape and pose corrective model; A facial blendshape model.Its design is friendly for both CG and CV communities.
picoclaw
Video to OpenSim motion data using SAM3D Body for 3D pose estimation
Monocular video to OpenSim motion data using RTMDet + RTMW + MotionBERT
Repositories
34Momentum Human Rig is an anatomically-inspired parametric full-body digital human model developed at Meta. It includes: A parametric body skeletal model; A realistic 3D mesh skinned to the skeleton with levels of detail;A body blendshape and pose corrective model; A facial blendshape model.Its design is friendly for both CG and CV communities.
Markerless kinematics with any cameras — From 2D Pose estimation to 3D OpenSim motion
picoclaw
Video to OpenSim motion data using SAM3D Body for 3D pose estimation
Monocular video to OpenSim motion data using RTMDet + RTMW + MotionBERT
Cross-platform, customizable ML solutions for live and streaming media.
No description provided.
The official Pytorch code for MoRo: Masked Modeling for Human Motion Recovery Under Occlusions
PromptHMR: Promptable Human Mesh Recovery
Anny, A Free and Interpretable Human Body Model for all ages, written in PyTorch.
Package to read, preprocess, vizualise, and synchronise soccer event and tracking data.
"DeepTutor: AI-Powered Personalized Learning Assistant"
🏂 Training-Free Human Mesh Recovery from Videos, based on SAM-3, Diffusion-VAS, and SAM-3D-Body.
Track-Anything is a flexible and interactive tool for video object tracking and segmentation, based on Segment Anything, XMem, and E2FGVI.
Perceive Anything: Recognize, Explain, Caption, and Segment Anything in Images and Videos
No description provided.
Gradio demo for the monocular fitting code
https://vaila.readthedocs.io/
The repository provides code for running inference and finetuning with the Meta Segment Anything Model 3 (SAM 3), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.
🐜🐀🐒🚶 A toolkit for robust markerless 3D pose estimation
No description provided.
Compute 2D human pose and angles from a video or a webcam.
Test
Source code of the marker enhancer
footPress is a MATLAB based toolbox which can be used for visualization and analysis of plantar pressure data.
A collection of Python scripts and applications for supporting quantitative behavioral observation using videos.
JARVIS Markerless 3D Motion Capture Pytorch Library
Multi-Joint dynamics with Contact. A general purpose physics simulator.
Dataset and Code for CVSports at CVPR 2025 paper "AthletePose3D: A Benchmark Dataset for 3D Human Pose Estimation and Kinematic Validation in Athletic Movements"
From a body shape, infer the anatomic skeleton.