Dataset Versions

v2

Version 2

Generated on May 23, 2024

Dataset Split

Train Set 87%
684Images
Valid Set 8%
65Images
Test Set 4%
33Images

Preprocessing

Auto-Orient: Applied
Static Crop: 25-75% Horizontal Region, 25-75% Vertical Region
Resize: Fit within 160x160

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Brightness: Between -15% and +15%