[Feature] Add gt_edge_map field.

mmseg/datasets/transforms/formatting.py

@@ -73,6 +73,12 @@ def transform(self, results: dict) -> dict:
  ...].astype(np.int64)))
  data_sample.gt_sem_seg = PixelData(**gt_sem_seg_data)
 
+ if 'gt_edge_map' in results:
+ gt_edge_data = dict(
+ data=to_tensor(results['gt_edge_map'][None,
+ ...].astype(np.int64)))
+ data_sample.set_data(dict(gt_edge_map=PixelData(**gt_edge_data)))
+
  img_meta = {}
  for key in self.meta_keys:
  if key in results:

mmseg/datasets/transforms/transforms.py

@@ -1245,7 +1245,7 @@ class GenerateEdge(BaseTransform):
  - gt_seg_map
 
  Added Keys:
- - gt_edge (np.ndarray, uint8): The edge annotation generated from the
+ - gt_edge_map (np.ndarray, uint8): The edge annotation generated from the
  seg map by extracting border between different semantics.
 
  Args:
@@ -1296,7 +1296,7 @@ def transform(self, results: Dict) -> Dict:
  (self.edge_width, self.edge_width))
  edge = cv2.dilate(edge, kernel)
 
- results['gt_edge'] = edge
+ results['gt_edge_map'] = edge
  results['edge_width'] = self.edge_width
 
  return results

mmseg/utils/misc.py

@@ -98,6 +98,11 @@ def stack_batch(inputs: List[torch.Tensor],
  del data_sample.gt_sem_seg.data
  data_sample.gt_sem_seg.data = F.pad(
  gt_sem_seg, padding_size, value=seg_pad_val)
+ if 'gt_edge_map' in data_sample:
+ gt_edge_map = data_sample.gt_edge_map.data
+ del data_sample.gt_edge_map.data
+ data_sample.gt_edge_map.data = F.pad(
+ gt_edge_map, padding_size, value=seg_pad_val)
  data_sample.set_metainfo({
  'img_shape': tensor.shape[-2:],
  'pad_shape': data_sample.gt_sem_seg.shape,

tests/test_datasets/test_transform.py

@@ -792,7 +792,7 @@ def test_generate_edge():
  results['img_shape'] = seg_map.shape
 
  results = transform(results)
- assert np.all(results['gt_edge'] == np.array([
+ assert np.all(results['gt_edge_map'] == np.array([
  [0, 0, 0, 1, 0],
  [0, 0, 1, 1, 1],
  [0, 1, 1, 1, 0],