Photo Type Classification Image Dataset

augmented-v1

Version 1
Generated Jan 29, 2022

Train / Test Split

Training Set

97
1.1k
images

Validation Set

2
20
images

Testing Set

1
10
images
Preprocessing
Auto-Orient: Applied
Resize: Fill (with center crop) in 416x416
Augmentations
Outputs per training example: 15
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Rotation: Between -5° and +5°
Grayscale: Apply to 5% of images
Hue: Between -10° and +10°
Exposure: Between -10% and +10%
Blur: Up to 1px
Details
Version Name: augmented-v1
Version ID: 1
Generated: Jan 29, 2022
Annotation Group: photos