DetectAndTrack
A CVPR 2018 algorithm for efficient multi-person pose estimation and tracking in videos, ranking first in the ICCV 2017 PoseTrack challenge.
What is DetectAndTrack?
DetectAndTrack is a research implementation from Facebook AI Research (FAIR) that provides an efficient algorithm for detecting and tracking human poses in videos. It solves the problem of multi-person pose estimation across video frames by combining detection and tracking into a single framework, which improves both accuracy and computational efficiency compared to separate systems.
Computer vision researchers and engineers working on video analysis, human pose estimation, activity recognition, or multi-object tracking, particularly those interested in reproducing or building upon state-of-the-art methods from academic literature.
Developers choose DetectAndTrack for its proven performance (it ranked first in the PoseTrack challenge), its efficient unified approach to detection and tracking, and the availability of pre-trained models and detailed configurations for both 2D and 3D models, which facilitate research and application development.
Overview
The implementation of an algorithm presented in the CVPR18 paper: "Detect-and-Track: Efficient Pose Estimation in Videos"
Use Cases
Best For
- Reproducing state-of-the-art pose estimation and tracking results from academic papers
Related Projects
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
