Dataset Versions

v4

2024-10-17 11:52pm

Generated on Oct 17, 2024

Dataset Split

Train Set 83%
2657Images
Valid Set 10%
327Images
Test Set 6%
204Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640
Filter Null: Require at least 63% of images to contain annotations.

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Rotation: Between -14° and +14°

Similar Projects

See More