Trial

Segmentation

Instance Segmentation

Segmentation Image Dataset

v4

2022-12-27 3:25pm

Generated on Dec 27, 2022

Dataset Split

Train Set 79%
30Images
Valid Set 5%
2Images
Test Set 16%
6Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 1080x1080

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Rotation: Between -6° and +6°
Shear: ±15° Horizontal, ±15° Vertical
Grayscale: Apply to 38% of images
Exposure: Between -41% and +41%
Blur: Up to 1px
Bounding Box: Flip: Horizontal, Vertical
Bounding Box: 90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Bounding Box: Crop: 0% Minimum Zoom, 20% Maximum Zoom
Bounding Box: Rotation: Between -15° and +15°
Bounding Box: Noise: Up to 2% of pixels