duran

obj

Semantic Segmentation

obj Image Dataset

v12

2023-03-22 9:34am

Generated on Mar 22, 2023

Dataset Split

Train Set 65%
2034Images
Valid Set 20%
621Images
Test Set 15%
468Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 2048x2048
Tile: 3 rows x 3 columns

Augmentations

Outputs per training example: 1
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Crop: 0% Minimum Zoom, 15% Maximum Zoom
Rotation: Between -30° and +30°
Shear: ±14° Horizontal, ±8° Vertical
Brightness: Between -5% and +0%