KuangJuiHsu/DeepCO3

[CVPR19] DeepCO3: Deep Instance Co-segmentation by Co-peak Search and Co-saliency (Oral paper)

31
/ 100
Emerging

This helps researchers in computer vision automatically identify and segment individual objects from a collection of related images, even if those objects appear multiple times. You input a set of images containing instances of the same object category, and it outputs a distinct segmentation mask for each individual object found. This is ideal for scientists working with large visual datasets who need precise object isolation without extensive manual annotation.

137 stars. No commits in the last 6 months.

Use this if you need to precisely outline every instance of a target object across a group of images, especially when you don't have pixel-level annotations for training.

Not ideal if you are looking to segment objects in a single image or if you have abundant pixel-wise labeled data for training a standard segmentation model.

computer-vision image-analysis object-segmentation visual-data-processing
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 8 / 25
Community 13 / 25

How are scores calculated?

Stars

137

Forks

13

Language

MATLAB

License

Last pushed

Apr 30, 2019

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/KuangJuiHsu/DeepCO3"

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