Limelight 2023 FRC Dataset Aggregator
Limelight OS 2023 supports learning-based vision with deep neural networks. We're crowdsourcing images of cubes, cones, and robots to build an object detection model for FRC robots!
Please upload images containing any combination of cubes, cones, and robots.
Your images will not be publicized.
Your images should be as broadly diverse as possible. They should include diverse backgrounds, lighting, environments, and perspectives.
Game pieces and robots should not be centered in all of your images.
Please feel free to include:
- Damaged and marked game pieces
- Blurry or out-of-focus images
- Rotated (+/- 45 degree) images
- Under-exposed and Over-exposed (dark and bright) images
- Covered / partially visible game pieces
- Messy environments
- Images that seem "difficult" to understand
- Non-standard environments such as outdoor environments, home environments, classrooms, wooden fields, CAD environments, etc. Get as creative as you want.