VerVelVel

things

Instance Segmentation

things Image Dataset

v1

2024-05-22 2:30pm

Generated on May 22, 2024

Dataset Split

Train Set 87%
222Images
Valid Set 8%
21Images
Test Set 4%
11Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 3
90° Rotate: Clockwise, Counter-Clockwise
Brightness: Between -24% and +24%
Noise: Up to 1.88% of pixels
Cutout: 5 boxes with 10% size each
incabinhuman
317 images
dog-cat-person
1381 images
cats
1206 images