HHousen/object-discovery-pytorch

An implementation of several unsupervised object discovery models (Slot Attention, SLATE, GNM) in PyTorch with pre-trained models.

37
/ 100
Emerging

This project helps machine learning researchers experiment with advanced computer vision models that can automatically identify and separate individual objects within complex images, even when they're overlapping or poorly defined. You provide a dataset of images, and the system outputs trained models capable of dissecting these images into their constituent parts, similar to how humans perceive objects. It's designed for researchers exploring unsupervised object recognition techniques.

No commits in the last 6 months.

Use this if you are a machine learning researcher who needs to train and evaluate state-of-the-art unsupervised object discovery models like Slot Attention, SLATE, or GNM on your own image datasets.

Not ideal if you need a plug-and-play solution for general object detection or segmentation with pre-defined categories, or if you lack a CUDA-enabled computing device and experience with Python development environments.

computer-vision unsupervised-learning image-analysis object-recognition machine-learning-research
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 5 / 25
Maturity 16 / 25
Community 14 / 25

How are scores calculated?

Stars

14

Forks

3

Language

Python

License

Apache-2.0

Last pushed

May 26, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/HHousen/object-discovery-pytorch"

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