Dataset Versions

v2

2024-01-05 2:26am

Generated on Jan 4, 2024

Dataset Split

Train Set 88%
381Images
Valid Set 7%
31Images
Test Set 5%
22Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 3
Saturation: Between -15% and +15%
Exposure: Between -14% and +14%
Cutout: 3 boxes with 4% size each
Mosaic: Applied
Bounding Box: Brightness: Between 0% and +20%
Bounding Box: Noise: Up to 0.69% of pixels