twke18/HSG

Unsupervised Hierarchical Semantic Segmentation with Multiview Cosegmentation and Clustering Transformers

35
/ 100
Emerging

This project helps computer vision researchers and practitioners automatically identify distinct objects and regions within images without requiring pre-labeled examples. It takes raw image data and outputs hierarchical segmentations, meaning it can group pixels into fine-grained details or broader categories, offering a flexible understanding of an image's content. This is for professionals working with image analysis, visual data interpretation, or autonomous systems.

No commits in the last 6 months.

Use this if you need to segment images into meaningful, multi-level groupings (from fine details to broad categories) without the time-consuming and expensive process of manually labeling thousands of images.

Not ideal if your segmentation task requires precise, pre-defined object classes that are consistent across all images, as this method discovers groupings rather than recognizing specific categories.

image-analysis unsupervised-learning object-recognition computer-vision scene-understanding
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 9 / 25
Maturity 16 / 25
Community 10 / 25

How are scores calculated?

Stars

74

Forks

6

Language

Python

License

MIT

Last pushed

Apr 02, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/twke18/HSG"

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