mkocabas/PARE
PyTorch implementation of PARE, a deep learning method for estimating 3D human body pose and shape from single images.
Not currently ranked — collecting fresh signals.
star history
PARE (Part Attention Regressor) is a computer vision method that estimates 3D human body pose and shape from 2D images using part-level attention mechanisms. Implemented in PyTorch, it handles occlusions robustly and includes demo and evaluation code. The project provides pre-trained models and scripts for running inference on video or image inputs.
Frequently asked
- What is mkocabas/PARE?
- PyTorch implementation of PARE, a deep learning method for estimating 3D human body pose and shape from single images.
- Is PARE open source?
- Yes — mkocabas/PARE is an open-source project tracked on heatdrop.
- What language is PARE written in?
- mkocabas/PARE is primarily written in Python.
- How popular is PARE?
- mkocabas/PARE has 416 stars on GitHub.
- Where can I find PARE?
- mkocabas/PARE is on GitHub at https://github.com/mkocabas/PARE.