Lane Area Semantic Segmentation Dataset

v4

Augmented Nov 27

Generated on Nov 27, 2022

Dataset Split

Train Set 88%
1836Images
Valid Set 7%
141Images
Test Set 5%
102Images

Preprocessing

Filter Null: Require all images to contain annotations.

Augmentations

Outputs per training example: 3
Flip: Horizontal
Crop: 0% Minimum Zoom, 35% Maximum Zoom
Rotation: Between -15° and +15°
Shear: ±21° Horizontal, ±15° Vertical
Noise: Up to 1% of pixels