Site segmentation

Mask R-CNN V1

• Version/date: 1 (April 3, 2025).
• Architecture: Mask R-CNN.
• Base model: COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml.
• Dataset: Sites V1.
• Training split: 80% training, 20% evaluation.
• Evaluation dataset: Eval data (provided by James).
• Model weights: mask_rcnn_dataset_v1_2025.pth.

Notes

• [ ] Need to fix the class name in the model (currently: "person" -> "site").
• [x] Improve and organize training scripts.
• [ ] Set up training monitoring.
• [x] Convert output masks into proper polygons (GeoJSON/KML).

Results

Performance metrics:

Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.397
Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.706
Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.414
Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.109
Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.427
Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.469
Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.568
Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.572
Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.002
Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.205
Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.611

Sample outputs:

Issues noticed:

• It only finds around 50% of the sites found - need to improve dataset.