eugeneteoh/greenaug

GreenAug: Green Screen Augmentation Enables Scene Generalisation in Robotic Manipulation

27
/ 100
Experimental

This project helps robotics engineers train robotic arms to perform manipulation tasks more reliably across different real-world environments. By taking images or videos of robots in a green screen setup, it intelligently replaces the background with diverse, realistic, or even imagined scenes. The result is a more robust robotic policy that can handle unexpected variations in its operational setting, making deployment easier and more effective.

No commits in the last 6 months.

Use this if you are a robotics engineer developing manipulation policies and want your robots to generalize better to various environments beyond their training setup.

Not ideal if you are looking for a general-purpose image augmentation tool unrelated to robotic manipulation, or if you don't use a green screen for your robot's visual data.

robotics training robotic manipulation sim-to-real computer vision reinforcement learning
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 16 / 25
Community 6 / 25

How are scores calculated?

Stars

12

Forks

1

Language

Python

License

MIT

Last pushed

Sep 10, 2024

Commits (30d)

0

Get this data via API

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

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