Dataset Versions
v1
2024-08-10 3:57am
Generated on Aug 10, 2024
1497 Total Images
View All Images
Dataset Split
Train Set 88%
1311Images
Valid Set 8%
125Images
Test Set 4%
61Images
Preprocessing
Auto-Orient: Applied
Resize: Fill (with center crop) in 640x640
Filter Null: Require at least 70% of images to contain annotations.
Augmentations
Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Grayscale: Apply to 14% of images
Saturation: Between -25% and +25%
Brightness: Between -20% and +20%
Exposure: Between -12% and +12%
Blur: Up to 1.3px