Emory

Vocal Flaps

Instance Segmentation

Vocal Flaps Image Dataset

v1

2023-06-07 11:34am

Generated on Jun 7, 2023

Dataset Split

Train Set 90%
72Images
Valid Set 6%
5Images
Test Set 4%
3Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
Rotation: Between -19° and +19°
Grayscale: Apply to 25% of images
Hue: Between -21° and +21°
Saturation: Between -51% and +51%
Blur: Up to 2px
Noise: Up to 4% of pixels