baegwangbin/surface_normal_uncertainty
[ICCV 2021 Oral] Estimating and Exploiting the Aleatoric Uncertainty in Surface Normal Estimation
This project helps computer vision practitioners analyze 3D scenes by providing accurate surface normal estimations from 2D images, even for complex or poorly lit objects. It takes an input image and outputs a map of surface normals along with a per-pixel uncertainty measure. Computer vision engineers and researchers working on 3D reconstruction, robotics, or augmented reality would find this useful.
245 stars. No commits in the last 6 months.
Use this if you need to reliably estimate the orientation of surfaces in an image, especially when dealing with varied scenes or objects where traditional methods might struggle.
Not ideal if you primarily need depth estimation rather than surface orientation, or if you require real-time processing on very constrained hardware without powerful GPUs.
Stars
245
Forks
21
Language
Python
License
MIT
Category
Last pushed
Aug 18, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/baegwangbin/surface_normal_uncertainty"
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.