monniert/dti-sprites

(ICCV 2021) Code for "Unsupervised Layered Image Decomposition into Object Prototypes" paper

40
/ 100
Emerging

This project helps computer vision researchers and AI practitioners analyze complex images by breaking them down into simpler, overlapping components, like separating a crowded scene into individual objects. It takes raw image data and outputs distinct 'sprite' images representing recurring objects or patterns, along with how they combine to form the original image. This is useful for anyone working on image understanding, object recognition, or generative AI models.

No commits in the last 6 months.

Use this if you need to automatically decompose images into their constituent parts without prior labeling, allowing you to understand or manipulate individual objects within a scene.

Not ideal if you require object detection with precise bounding box coordinates or have a readily available dataset with labeled objects.

computer-vision image-segmentation object-recognition unsupervised-learning generative-ai
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 16 / 25

How are scores calculated?

Stars

46

Forks

8

Language

Python

License

MIT

Last pushed

Feb 01, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/monniert/dti-sprites"

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