ksr

Edit Project

pt-detect

Instance Segmentation

pt-detect Image Dataset

Try Pre-Trained Model
Versions
2023-03-27 4:43am
v15
Mar 27, 2023
2023-02-09 12:09pm
v14
Feb 9, 2023
2023-02-06 4:03pm
v13
Feb 6, 2023
2023-02-04 9:49am
v12
Feb 4, 2023
2023-01-30 12:15pm
v11
Jan 30, 2023
2023-01-27 11:34am
v10
Jan 27, 2023
2023-01-11 5:19pm
v9
Jan 11, 2023
2023-01-09 5:46pm
v8
Jan 9, 2023
2023-01-09 10:02am
v7
Jan 9, 2023
2023-01-06 12:24pm
v6
Jan 6, 2023
2023-01-06 10:24am
v5
Jan 6, 2023
2023-01-06 10:12am
v4
Jan 6, 2023
2023-01-06 10:00am
v3
Jan 6, 2023
2023-01-04 3:45pm
v2
Jan 4, 2023
2023-01-04 2:03pm
v1
Jan 4, 2023
v15

2023-03-27 4:43am

Generated on Mar 27, 2023
Download Dataset

5283 Total Images

View All Images
Annotation VisualizationDataset Image Annotation VisualizationDataset Image Annotation VisualizationDataset Image Annotation VisualizationDataset Image Annotation VisualizationDataset Image Annotation VisualizationDataset Image Annotation VisualizationDataset Image Annotation VisualizationDataset Image Annotation VisualizationDataset Image Annotation VisualizationDataset Image Annotation VisualizationDataset Image Annotation VisualizationDataset Image

Dataset Split

Train Set 92%
4866Images
Valid Set 6%
302Images
Test Set 2%
115Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640
Modify Classes: 1 remapped, 0 dropped

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Rotation: Between -23° and +23°
Shear: ±20° Horizontal, ±15° Vertical
Grayscale: Apply to 28% of images
Brightness: Between 0% and +38%
Noise: Up to 3% of pixels