tohinz/multiple-objects-gan

Implementation for "Generating Multiple Objects at Spatially Distinct Locations" (ICLR 2019)

39
/ 100
Emerging

This project helps researchers in computer vision generate images containing multiple distinct objects, positioned accurately. You provide existing datasets of images like MNIST digits or objects from the CLEVR and MS-COCO datasets, and it produces new, synthesized images where the objects are correctly placed and structured according to the trained model. This tool is for researchers and students working on generative models and synthetic image creation.

112 stars. No commits in the last 6 months.

Use this if you need to create synthetic images containing multiple objects arranged in specific, learned spatial configurations for research or experimental purposes.

Not ideal if you're looking for a simple, out-of-the-box image generation tool for everyday use or commercial applications, as it requires familiarity with research setups and specific datasets.

generative-AI image-synthesis computer-vision-research synthetic-data-generation deep-learning-experiments
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 9 / 25
Maturity 16 / 25
Community 14 / 25

How are scores calculated?

Stars

112

Forks

14

Language

Python

License

MIT

Category

gan-based-t2i

Last pushed

Jan 13, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/tohinz/multiple-objects-gan"

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