LarsDoorenbos/ccdm-stochastic-segmentation

Repository for "Stochastic Segmentation with Conditional Categorical Diffusion Models" (ICCV 2023)

39
/ 100
Emerging

This project helps medical professionals and autonomous driving engineers by providing a tool to generate multiple possible segmentation maps for an image, accounting for situations where there isn't a single 'correct' answer. You input an image, and it outputs several potential label maps, reflecting different plausible interpretations. It's designed for anyone who needs to understand the inherent uncertainty in image segmentation, especially in critical applications.

No commits in the last 6 months.

Use this if you need to understand the range of possible segmentations for an image, rather than just one definitive outline, particularly in fields like medical imaging or autonomous vehicle perception.

Not ideal if your application strictly requires a single, deterministic segmentation map with no ambiguity or if you are looking for a simple, off-the-shelf image segmentation tool for non-critical tasks.

medical-diagnostics autonomous-driving image-segmentation uncertainty-quantification radiology
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 15 / 25

How are scores calculated?

Stars

52

Forks

8

Language

Python

License

MIT

Last pushed

Oct 09, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/LarsDoorenbos/ccdm-stochastic-segmentation"

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