wpeebles/gangealing

Official PyTorch Implementation of "GAN-Supervised Dense Visual Alignment" (CVPR 2022 Oral, Best Paper Finalist)

46
/ 100
Emerging

This project helps anyone working with collections of images or videos by automatically aligning them to a common template, even if they show the same object from different angles or poses. You provide a set of unaligned images (like different cats) and it outputs a 'congealed' average template along with the tools to propagate details from that template to any other image or video of that object. This is useful for animators, designers, or researchers who need to apply consistent visual elements across varying visual data.

1,013 stars. No commits in the last 6 months.

Use this if you need to precisely align diverse images or video frames of a specific object category (like different cats, cars, or human faces) to a standardized reference.

Not ideal if you're looking for general image comparison or object detection, as this tool focuses specifically on dense visual alignment and propagating changes.

animation image-processing visual-effects design computer-vision-research
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 20 / 25

How are scores calculated?

Stars

1,013

Forks

121

Language

Python

License

BSD-2-Clause

Last pushed

Oct 12, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/wpeebles/gangealing"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.