FangjinhuaWang/PatchmatchNet
Official code of PatchmatchNet (CVPR 2021 Oral)
This project helps specialists like surveyors, architects, or cultural heritage preservationists convert multiple 2D photos of a scene into a detailed 3D point cloud model. You provide a set of images taken from different angles, along with their camera settings, and it generates a textured 3D representation. It's designed for users who need to reconstruct complex environments or objects with high accuracy.
548 stars. No commits in the last 6 months.
Use this if you need to create accurate 3D models from multiple photographs, especially for high-resolution images, and are comfortable working with structured datasets and command-line tools.
Not ideal if you're looking for a simple, automated 'point-and-shoot' 3D reconstruction tool without needing to manage camera parameters or dataset organization.
Stars
548
Forks
74
Language
Python
License
MIT
Category
Last pushed
Sep 19, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/FangjinhuaWang/PatchmatchNet"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
3DOM-FBK/deep-image-matching
Multiview matching with deep-learning and hand-crafted local features for COLMAP and other SfM...
suhangpro/mvcnn
Multi-view CNN (MVCNN) for shape recognition
zouchuhang/LayoutNet
Torch implementation of our CVPR 18 paper: "LayoutNet: Reconstructing the 3D Room Layout from a...
andyzeng/tsdf-fusion-python
Python code to fuse multiple RGB-D images into a TSDF voxel volume.
andyzeng/tsdf-fusion
Fuse multiple depth frames into a TSDF voxel volume.