shahaf1313/ProCST

Official implementation of ProCST image-to-image translation for UDA-SS.

32
/ 100
Emerging

This project helps computer vision researchers and developers improve the accuracy of semantic segmentation models when applying them to new visual environments. It takes synthetic images (like from video games) and transforms their visual style to closely resemble real-world images, such as urban street scenes. The output is a set of 'style-adapted' images that can then be used with standard unsupervised domain adaptation techniques to achieve better segmentation results in the real-world target domain.

No commits in the last 6 months.

Use this if you need to bridge the visual gap between a simulated or synthetic dataset and a real-world target dataset for semantic segmentation tasks.

Not ideal if your image segmentation problem does not involve adapting models from a source domain to a visually different target domain, or if you are not working with semantic segmentation.

computer-vision semantic-segmentation unsupervised-domain-adaptation image-style-transfer AI-model-training
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 7 / 25
Maturity 16 / 25
Community 9 / 25

How are scores calculated?

Stars

29

Forks

3

Language

Python

License

MIT

Last pushed

Jan 14, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/shahaf1313/ProCST"

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