Dataset Versions

v8

2024-04-07 1:40pm

Generated on Apr 7, 2024

Dataset Split

Train Set 88%
153Images
Valid Set 8%
13Images
Test Set 4%
7Images

Preprocessing

Auto-Orient: Applied
Resize: Fill (with center crop) in 1920x1080

Augmentations

Outputs per training example: 3
Flip: Horizontal
Crop: 0% Minimum Zoom, 30% Maximum Zoom
Rotation: Between -15° and +15°
Shear: ±10° Horizontal, ±10° Vertical
Hue: Between -25° and +25°
Saturation: Between -25% and +25%
Brightness: Between -25% and +25%
Exposure: Between -10% and +10%
Blur: Up to 2.5px
Noise: Up to 0.1% of pixels