[Feature]Generate Edge for dataset (#2210)

* [WIP]Generate Edge for dataset

* add ut

* add repr

* add inti

mmseg/datasets/__init__.py

@@ -15,7 +15,7 @@ from .night_driving import NightDrivingDataset
 from .pascal_context import PascalContextDataset, PascalContextDataset59
 from .potsdam import PotsdamDataset
 from .stare import STAREDataset
-from .transforms import (CLAHE, AdjustGamma, LoadAnnotations,
+from .transforms import (CLAHE, AdjustGamma, GenerateEdge, LoadAnnotations,
                          LoadBiomedicalAnnotation, LoadBiomedicalData,
                          LoadBiomedicalImageFromFile, LoadImageFromNDArray,
                          PackSegInputs, PhotoMetricDistortion, RandomCrop,
@@ -33,5 +33,5 @@ __all__ = [
     'RandomRotate', 'AdjustGamma', 'CLAHE', 'Rerange', 'RGB2Gray',
     'RandomCutOut', 'RandomMosaic', 'PackSegInputs', 'ResizeToMultiple',
     'LoadImageFromNDArray', 'LoadBiomedicalImageFromFile',
-    'LoadBiomedicalAnnotation', 'LoadBiomedicalData'
+    'LoadBiomedicalAnnotation', 'LoadBiomedicalData', 'GenerateEdge'
 ]

mmseg/datasets/transforms/__init__.py

@@ -3,14 +3,15 @@ from .formatting import PackSegInputs
 from .loading import (LoadAnnotations, LoadBiomedicalAnnotation,
                       LoadBiomedicalData, LoadBiomedicalImageFromFile,
                       LoadImageFromNDArray)
-from .transforms import (CLAHE, AdjustGamma, PhotoMetricDistortion, RandomCrop,
+from .transforms import (CLAHE, AdjustGamma, GenerateEdge,
-                         RandomCutOut, RandomMosaic, RandomRotate, Rerange,
+                         PhotoMetricDistortion, RandomCrop, RandomCutOut,
-                         ResizeToMultiple, RGB2Gray, SegRescale)
+                         RandomMosaic, RandomRotate, Rerange, ResizeToMultiple,
+                         RGB2Gray, SegRescale)
 
 __all__ = [
     'LoadAnnotations', 'RandomCrop', 'SegRescale', 'PhotoMetricDistortion',
     'RandomRotate', 'AdjustGamma', 'CLAHE', 'Rerange', 'RGB2Gray',
     'RandomCutOut', 'RandomMosaic', 'PackSegInputs', 'ResizeToMultiple',
     'LoadImageFromNDArray', 'LoadBiomedicalImageFromFile',
-    'LoadBiomedicalAnnotation', 'LoadBiomedicalData'
+    'LoadBiomedicalAnnotation', 'LoadBiomedicalData', 'GenerateEdge'
 ]

mmseg/datasets/transforms/transforms.py

@@ -1,7 +1,8 @@
 # Copyright (c) OpenMMLab. All rights reserved.
 import copy
-from typing import Sequence, Tuple, Union
+from typing import Dict, Sequence, Tuple, Union
 
+import cv2
 import mmcv
 import numpy as np
 from mmcv.transforms.base import BaseTransform
@@ -1147,3 +1148,81 @@ class RandomMosaic(BaseTransform):
         repr_str += f'pad_val={self.pad_val}, '
         repr_str += f'seg_pad_val={self.pad_val})'
         return repr_str
+
+
+@TRANSFORMS.register_module()
+class GenerateEdge(BaseTransform):
+    """Generate Edge for CE2P approach.
+
+    Edge will be used to calculate loss of
+    `CE2P <https://arxiv.org/abs/1809.05996>`_.
+
+    Modified from https://github.com/liutinglt/CE2P/blob/master/dataset/target_generation.py # noqa:E501
+
+    Required Keys:
+
+        - img_shape
+        - gt_seg_map
+
+    Added Keys:
+        - gt_edge (np.ndarray, uint8): The edge annotation generated from the
+            seg map by extracting border between different semantics.
+
+    Args:
+        edge_width (int): The width of edge. Default to 3.
+        ignore_index (int): Index that will be ignored. Default to 255.
+    """
+
+    def __init__(self, edge_width: int = 3, ignore_index: int = 255) -> None:
+        super().__init__()
+        self.edge_width = edge_width
+        self.ignore_index = ignore_index
+
+    def transform(self, results: Dict) -> Dict:
+        """Call function to generate edge from segmentation map.
+
+        Args:
+            results (dict): Result dict.
+
+        Returns:
+            dict: Result dict with edge mask.
+        """
+        h, w = results['img_shape']
+        edge = np.zeros((h, w), dtype=np.uint8)
+        seg_map = results['gt_seg_map']
+
+        # down
+        edge_down = edge[1:h, :]
+        edge_down[(seg_map[1:h, :] != seg_map[:h - 1, :])
+                  & (seg_map[1:h, :] != self.ignore_index) &
+                  (seg_map[:h - 1, :] != self.ignore_index)] = 1
+        # left
+        edge_left = edge[:, :w - 1]
+        edge_left[(seg_map[:, :w - 1] != seg_map[:, 1:w])
+                  & (seg_map[:, :w - 1] != self.ignore_index) &
+                  (seg_map[:, 1:w] != self.ignore_index)] = 1
+        # up_left
+        edge_upleft = edge[:h - 1, :w - 1]
+        edge_upleft[(seg_map[:h - 1, :w - 1] != seg_map[1:h, 1:w])
+                    & (seg_map[:h - 1, :w - 1] != self.ignore_index) &
+                    (seg_map[1:h, 1:w] != self.ignore_index)] = 1
+        # up_right
+        edge_upright = edge[:h - 1, 1:w]
+        edge_upright[(seg_map[:h - 1, 1:w] != seg_map[1:h, :w - 1])
+                     & (seg_map[:h - 1, 1:w] != self.ignore_index) &
+                     (seg_map[1:h, :w - 1] != self.ignore_index)] = 1
+
+        kernel = cv2.getStructuringElement(cv2.MORPH_RECT,
+                                           (self.edge_width, self.edge_width))
+        edge = cv2.dilate(edge, kernel)
+
+        results['gt_edge'] = edge
+        results['edge_width'] = self.edge_width
+
+        return results
+
+    def __repr__(self):
+        repr_str = self.__class__.__name__
+        repr_str += f'edge_width={self.edge_width}, '
+        repr_str += f'ignore_index={self.ignore_index})'
+        return repr_str

tests/test_datasets/test_transform.py

@@ -679,3 +679,30 @@ def test_resize_to_multiple():
     assert results['img'].shape == (224, 256, 3)
     assert results['gt_semantic_seg'].shape == (224, 256)
     assert results['img_shape'] == (224, 256)
+
+
+def test_generate_edge():
+    transform = dict(type='GenerateEdge', edge_width=1)
+    transform = TRANSFORMS.build(transform)
+
+    seg_map = np.array([
+        [1, 1, 1, 1, 1],
+        [1, 1, 1, 1, 2],
+        [1, 1, 1, 2, 2],
+        [1, 1, 2, 2, 2],
+        [1, 2, 2, 2, 2],
+        [2, 2, 2, 2, 2],
+    ])
+    results = dict()
+    results['gt_seg_map'] = seg_map
+    results['img_shape'] = seg_map.shape
+
+    results = transform(results)
+    assert np.all(results['gt_edge'] == np.array([
+        [0, 0, 0, 1, 0],
+        [0, 0, 1, 1, 1],
+        [0, 1, 1, 1, 0],
+        [1, 1, 1, 0, 0],
+        [1, 1, 0, 0, 0],
+        [1, 0, 0, 0, 0],
+    ]))