Fix resize when size == small_edge_size and max_size isn't None (#5409)

* Fix resize when size == small_edge_size and max_size isn't None

* Better test name

test/test_transforms.py

@@ -440,6 +440,19 @@ def test_resize_antialias_error():
         t(img)
 
 
+@pytest.mark.parametrize("height, width", ((32, 64), (64, 32)))
+def test_resize_size_equals_small_edge_size(height, width):
+    # Non-regression test for https://github.com/pytorch/vision/issues/5405
+    # max_size used to be ignored if size == small_edge_size
+    max_size = 40
+    img = Image.new("RGB", size=(width, height), color=127)
+
+    small_edge = min(height, width)
+    t = transforms.Resize(small_edge, max_size=max_size)
+    result = t(img)
+    assert max(result.size) == max_size
+
+
 class TestPad:
     def test_pad(self):
         height = random.randint(10, 32) * 2

torchvision/transforms/functional_pil.py

@@ -240,9 +240,6 @@ def resize(
         w, h = img.size
 
         short, long = (w, h) if w <= h else (h, w)
-        if short == size:
-            return img
-
         new_short, new_long = size, int(size * long / short)
 
         if max_size is not None:
@@ -255,6 +252,10 @@ def resize(
                 new_short, new_long = int(max_size * new_short / new_long), max_size
 
         new_w, new_h = (new_short, new_long) if w <= h else (new_long, new_short)
+
+        if (w, h) == (new_w, new_h):
+            return img
+        else:
             return img.resize((new_w, new_h), interpolation)
     else:
         if max_size is not None:

torchvision/transforms/functional_tensor.py

@@ -457,9 +457,6 @@ def resize(
         short, long = (w, h) if w <= h else (h, w)
         requested_new_short = size if isinstance(size, int) else size[0]
 
-        if short == requested_new_short:
-            return img
-
         new_short, new_long = requested_new_short, int(requested_new_short * long / short)
 
         if max_size is not None:
@@ -473,6 +470,9 @@ def resize(
 
         new_w, new_h = (new_short, new_long) if w <= h else (new_long, new_short)
 
+        if (w, h) == (new_w, new_h):
+            return img
+
     else:  # specified both h and w
         new_w, new_h = size[1], size[0]