Idan Kideckel

pools

Instance Segmentation

pools Image Dataset

v5

2023-06-01 12:18pm

Generated on Jun 1, 2023

Dataset Split

Train Set 80%
600Images
Valid Set 11%
83Images
Test Set 9%
67Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 3
Grayscale: Apply to 25% of images
Hue: Between -79° and +79°
Noise: Up to 8% of pixels
Cutout: 10 boxes with 10% size each