Encode segmentation label images as pascal classes Args: mask (np.ndarray): raw segmentation label image of dimension (M, N, 3), in which the Pascal classes are encoded as colours. Returns: (np.ndarray): class map with dimensions (M,N), where the value at a
(mask)
| 81 | |
| 82 | |
| 83 | def encode_segmap(mask): |
| 84 | """Encode segmentation label images as pascal classes |
| 85 | Args: |
| 86 | mask (np.ndarray): raw segmentation label image of dimension |
| 87 | (M, N, 3), in which the Pascal classes are encoded as colours. |
| 88 | Returns: |
| 89 | (np.ndarray): class map with dimensions (M,N), where the value at |
| 90 | a given location is the integer denoting the class index. |
| 91 | """ |
| 92 | mask = mask.astype(int) |
| 93 | label_mask = np.zeros((mask.shape[0], mask.shape[1]), dtype=np.int16) |
| 94 | for ii, label in enumerate(get_pascal_labels()): |
| 95 | label_mask[np.where(np.all(mask == label, axis=-1))[:2]] = ii |
| 96 | label_mask = label_mask.astype(int) |
| 97 | return label_mask |
| 98 | |
| 99 | |
| 100 | def decode_seg_map_sequence(label_masks, dataset="pascal"): |
nothing calls this directly
no test coverage detected