amazon-science/unified-ept

A Unified Efficient Pyramid Transformer for Semantic Segmentation, ICCVW 2021

40
/ 100
Emerging

This project helps computer vision researchers and practitioners accurately outline and classify different objects or regions within images, a process known as semantic segmentation. It takes an input image and outputs a map where each pixel is labeled with its corresponding object class, such as 'person,' 'car,' or 'road.' This is particularly useful for those working on tasks requiring precise object recognition and scene understanding.

No commits in the last 6 months.

Use this if you need to precisely segment objects and regions in images, especially when working with complex scenes or large datasets like ADE20k or PASCAL-Context.

Not ideal if you are looking for a simple, out-of-the-box solution without deep learning infrastructure, as it requires specific hardware and multiple library installations.

computer-vision image-analysis object-recognition scene-understanding image-segmentation
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 7 / 25
Maturity 16 / 25
Community 17 / 25

How are scores calculated?

Stars

31

Forks

8

Language

Python

License

Apache-2.0

Last pushed

Oct 11, 2021

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/amazon-science/unified-ept"

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