thib Laigre

TL_003

Instance Segmentation

TL_003 Image Dataset

v2

2023-10-06 2:37pm

Generated on Oct 6, 2023

Dataset Split

Train Set 88%
816Images
Valid Set 9%
80Images
Test Set 3%
32Images

Preprocessing

Auto-Adjust Contrast: Using Contrast Stretching
Tile: 2 rows x 2 columns

Augmentations

Outputs per training example: 3
Flip: Horizontal
Crop: 0% Minimum Zoom, 20% Maximum Zoom
Rotation: Between -19° and +19°
Shear: ±15° Horizontal, ±15° Vertical
Saturation: Between -25% and +25%
Brightness: Between -10% and +10%
Exposure: Between -10% and +10%