Dataset Versions

v4

test

Generated on May 10, 2022

Dataset Split

Train Set 97%
2805Images
Valid Set 2%
54Images
Test Set 1%
27Images

Preprocessing

Auto-Orient: Applied
Resize: Fill (with center crop) in 416x416

Augmentations

Outputs per training example: 15
Flip: Horizontal
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Rotation: Between -5° and +5°
Grayscale: Apply to 5% of images
Hue: Between -10° and +10°
Exposure: Between -10% and +10%
Blur: Up to 1px

Similar Projects

See More