NVlabs/blade
Official PyTorch implementation of BLADE: Single-view Body Mesh Estimation through Accurate Depth Estimation (CVPR 2025). BLADE tackles close-range human mesh recovery where perspective distortion is strongest, and solves for camera pose and focal length in addition to SMPL(-X) parameters.
This project helps capture precise 3D human body shapes and movements from a single camera image, even when the person is very close to the camera. It takes an image of a person and outputs a detailed 3D mesh model of their body, along with accurate information about their position and the camera's perspective. It is designed for professionals in animation, virtual reality, motion capture, or sports analysis who need accurate 3D human models from standard video.
Use this if you need highly accurate 3D human body models and poses reconstructed from single images, especially in scenarios where subjects are close to the camera, leading to significant perspective distortion.
Not ideal if your primary goal is general object detection or if you only need rough 2D pose estimations, as this tool is specifically optimized for detailed 3D human body reconstruction.
Stars
43
Forks
2
Language
Python
License
—
Category
Last pushed
Nov 04, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/NVlabs/blade"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
DeepLabCut/DeepLabCut
Official implementation of DeepLabCut: Markerless pose estimation of user-defined features with...
openpifpaf/openpifpaf
Official implementation of "OpenPifPaf: Composite Fields for Semantic Keypoint Detection and...
lambdaloop/anipose
🐜🐀🐒🚶 A toolkit for robust markerless 3D pose estimation
DIYer22/bpycv
Computer vision utils for Blender (generate instance annoatation, depth and 6D pose by one line code)
NeLy-EPFL/DeepFly3D
Motion capture (markerless 3D pose estimation) pipeline and helper GUI for tethered Drosophila.