Dataset Versions

v1

2024-01-20 12:25pm

Generated on Jan 20, 2024

Dataset Split

Train Set 93%
6765Images
Valid Set 4%
323Images
Test Set 2%
159Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
Rotation: Between -43° and +43°
Shear: ±16° Horizontal, ±16° Vertical
Blur: Up to 3.3px
Noise: Up to 5.09% of pixels
Cutout: 3 boxes with 15% size each