[Feature] Add `min_size` arg in `Resize` to keep the shape after resize bigger than slide window (#1318)

* [Feature] add setr_resize

* fix a bug

* fix

* fix

* fix

* fix

* fix

* fix

* fix

* fix

* fix

* fix

* fix

* fix

* fix

mmseg/datasets/pipelines/transforms.py

@@ -99,13 +99,19 @@ class Resize(object):
             Default: None
         keep_ratio (bool): Whether to keep the aspect ratio when resizing the
             image. Default: True
+        min_size (int, optional): The minimum size for input and the shape
+            of the image and seg map will not be less than ``min_size``.
+            As the shape of model input is fixed like 'SETR' and 'BEiT'.
+            Following the setting in these models, resized images must be
+            bigger than the crop size in ``slide_inference``. Default: None
     """
 
     def __init__(self,
                  img_scale=None,
                  multiscale_mode='range',
                  ratio_range=None,
-                 keep_ratio=True):
+                 keep_ratio=True,
+                 min_size=None):
         if img_scale is None:
             self.img_scale = None
         else:
@@ -126,6 +132,7 @@ class Resize(object):
         self.multiscale_mode = multiscale_mode
         self.ratio_range = ratio_range
         self.keep_ratio = keep_ratio
+        self.min_size = min_size
 
     @staticmethod
     def random_select(img_scales):
@@ -240,6 +247,23 @@ class Resize(object):
     def _resize_img(self, results):
         """Resize images with ``results['scale']``."""
         if self.keep_ratio:
+            if self.min_size is not None:
+                # TODO: Now 'min_size' is an 'int' which means the minimum
+                # shape of images is (min_size, min_size, 3). 'min_size'
+                # with tuple type will be supported, i.e. the width and
+                # height are not equal.
+                if min(results['scale']) < self.min_size:
+                    new_short = self.min_size
+                else:
+                    new_short = min(results['scale'])
+
+                h, w = results['img'].shape[:2]
+                if h > w:
+                    new_h, new_w = new_short * h / w, new_short
+                else:
+                    new_h, new_w = new_short, new_short * w / h
+                results['scale'] = (new_h, new_w)
+
             img, scale_factor = mmcv.imrescale(
                 results['img'], results['scale'], return_scale=True)
             # the w_scale and h_scale has minor difference

tests/test_data/test_transform.py

@@ -123,6 +123,31 @@ def test_resize():
     assert int(288 * 0.5) <= resized_results['img_shape'][0] <= 288 * 2.0
     assert int(512 * 0.5) <= resized_results['img_shape'][1] <= 512 * 2.0
 
+    # test min_size=640
+    transform = dict(type='Resize', img_scale=(2560, 640), min_size=640)
+    resize_module = build_from_cfg(transform, PIPELINES)
+    resized_results = resize_module(results.copy())
+    assert resized_results['img_shape'] == (640, 1138, 3)
+
+    # test min_size=640 and img_scale=(512, 640)
+    transform = dict(type='Resize', img_scale=(512, 640), min_size=640)
+    resize_module = build_from_cfg(transform, PIPELINES)
+    resized_results = resize_module(results.copy())
+    assert resized_results['img_shape'] == (640, 1138, 3)
+
+    # test h > w
+    img = np.random.randn(512, 288, 3)
+    results['img'] = img
+    results['img_shape'] = img.shape
+    results['ori_shape'] = img.shape
+    # Set initial values for default meta_keys
+    results['pad_shape'] = img.shape
+    results['scale_factor'] = 1.0
+    transform = dict(type='Resize', img_scale=(2560, 640), min_size=640)
+    resize_module = build_from_cfg(transform, PIPELINES)
+    resized_results = resize_module(results.copy())
+    assert resized_results['img_shape'] == (1138, 640, 3)
+
 
 def test_flip():
     # test assertion for invalid prob