semantic_segmentation

Semantic Segmentation Example

Annotation

labelme data_annotated --labels labels.txt --nodata --validatelabel exact --config '{shift_auto_shape_color: -2}'

Convert to VOC-format Dataset

# It generates:
#   - data_dataset_voc/JPEGImages
#   - data_dataset_voc/SegmentationClass
#   - data_dataset_voc/SegmentationClassNpy
#   - data_dataset_voc/SegmentationClassVisualization
./labelme2voc.py data_annotated data_dataset_voc --labels labels.txt --noobject

Fig 1. JPEG image (left), PNG label (center), JPEG label visualization (right)

Note that the label file contains only very low label values (ex. 0, 4, 14), and 255 indicates the __ignore__ label value (-1 in the npy file).
You can see the label PNG file by following.

labelme_draw_label_png data_dataset_voc/SegmentationClass/2011_000003.png