Dataset Versions

v2

2024-12-27 3:02pm

Generated on Dec 27, 2024

Dataset Split

Train Set 82%
798Images
Valid Set 9%
89Images
Test Set 9%
89Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Hue: Between -20° and +20°
Saturation: Between -25% and +25%
Brightness: Between -20% and +20%
Exposure: Between -15% and +15%
Blur: Up to 3px