NeurAI-Lab/biHomE
This is the official repo for the CVPR 2021 IMW paper: "Perceptual Loss for Robust Unsupervised Homography Estimation"
This project helps computer vision researchers and practitioners accurately align images, even when illumination or viewpoints change significantly. It takes two images of the same scene, possibly captured under different conditions or angles, and outputs the mathematical transformation (homography) needed to perfectly overlap them. This is useful for tasks like creating panoramas, video stabilization, or image stitching.
No commits in the last 6 months.
Use this if you need to precisely align images that might have drastic differences in lighting or viewing angles, and existing methods struggle with robustness.
Not ideal if you're dealing with images that are fundamentally different or don't represent the same scene, as it's designed for aligning related views.
Stars
38
Forks
5
Language
Python
License
MIT
Category
Last pushed
Jan 31, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/NeurAI-Lab/biHomE"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
drprojects/superpoint_transformer
Official PyTorch implementation of Superpoint Transformer introduced in [ICCV'23] "Efficient 3D...
yuxumin/PoinTr
[ICCV 2021 Oral] PoinTr: Diverse Point Cloud Completion with Geometry-Aware Transformers
charlesq34/frustum-pointnets
Frustum PointNets for 3D Object Detection from RGB-D Data
drprojects/DeepViewAgg
[CVPR'22 Best Paper Finalist] Official PyTorch implementation of the method presented in...
facebookresearch/votenet
Deep Hough Voting for 3D Object Detection in Point Clouds