YOLO format without normalization and denormalization (#924)

* Convert to and from YOLO format without normalization and denormalization to prevent rounding errors

* Fix yolo tests

albumentations/augmentations/bbox_utils.py

@@ -233,20 +233,23 @@ def convert_bbox_to_albumentations(bbox, source_format, rows, cols, check_validi
         # https://github.com/pjreddie/darknet/blob/f6d861736038da22c9eb0739dca84003c5a5e275/scripts/voc_label.py#L12
         bbox, tail = bbox[:4], tuple(bbox[4:])
         _bbox = np.array(bbox[:4])
-        if np.any((_bbox <= 0) | (_bbox > 1)):
-            raise ValueError("In YOLO format all labels must be float and in range (0, 1]")
+        if check_validity and np.any((_bbox <= 0) | (_bbox > 1)):
+            raise ValueError("In YOLO format all coordinates must be float and in range (0, 1]")
 
-        x, y, width, height = denormalize_bbox(bbox, rows, cols)
+        x, y, w, h = bbox
 
-        x_min = int(x - width / 2 + 1)
-        x_max = int(x_min + width)
-        y_min = int(y - height / 2 + 1)
-        y_max = int(y_min + height)
+        w_half, h_half = w / 2, h / 2
+        x_min = x - w_half
+        y_min = y - h_half
+        x_max = x_min + w
+        y_max = y_min + h
     else:
         (x_min, y_min, x_max, y_max), tail = bbox[:4], tuple(bbox[4:])
 
     bbox = (x_min, y_min, x_max, y_max) + tail
-    bbox = normalize_bbox(bbox, rows, cols)
+
+    if source_format != "yolo":
+        bbox = normalize_bbox(bbox, rows, cols)
     if check_validity:
         check_bbox(bbox)
     return bbox
@@ -280,20 +283,21 @@ def convert_bbox_from_albumentations(bbox, target_format, rows, cols, check_vali
         )
     if check_validity:
         check_bbox(bbox)
-    bbox = denormalize_bbox(bbox, rows, cols)
+
+    if target_format != "yolo":
+        bbox = denormalize_bbox(bbox, rows, cols)
     if target_format == "coco":
         (x_min, y_min, x_max, y_max), tail = bbox[:4], tuple(bbox[4:])
         width = x_max - x_min
         height = y_max - y_min
         bbox = (x_min, y_min, width, height) + tail
     elif target_format == "yolo":
-        # https://github.com/pjreddie/darknet/blob/f6d861736038da22c9eb0739dca84003c5a5e275/scripts/voc_label.py#L12
         (x_min, y_min, x_max, y_max), tail = bbox[:4], bbox[4:]
-        x = int((x_min + x_max) / 2 - 1)
-        y = int((y_min + y_max) / 2 - 1)
-        width = x_max - x_min
-        height = y_max - y_min
-        bbox = normalize_bbox((x, y, width, height) + tail, rows, cols)
+        x = (x_min + x_max) / 2.0
+        y = (y_min + y_max) / 2.0
+        w = x_max - x_min
+        h = y_max - y_min
+        bbox = (x, y, w, h) + tail
     return bbox
 
 

tests/test_bbox.py

@@ -77,8 +77,11 @@ def test_calculate_bbox_area(bbox, rows, cols, expected):
         ((20, 30, 40, 50, 99), "coco", (0.2, 0.3, 0.6, 0.8, 99)),
         ((20, 30, 60, 80), "pascal_voc", (0.2, 0.3, 0.6, 0.8)),
         ((20, 30, 60, 80, 99), "pascal_voc", (0.2, 0.3, 0.6, 0.8, 99)),
-        ((0.2, 0.3, 0.4, 0.5), "yolo", (0.01, 0.06, 0.41, 0.56)),
-        ((0.2, 0.3, 0.4, 0.5, 99), "yolo", (0.01, 0.06, 0.41, 0.56, 99)),
+        ((0.2, 0.3, 0.4, 0.5), "yolo", (0.00, 0.05, 0.40, 0.55)),
+        ((0.2, 0.3, 0.4, 0.5, 99), "yolo", (0.00, 0.05, 0.40, 0.55, 99)),
+        ((0.1, 0.1, 0.2, 0.2), "yolo", (0.0, 0.0, 0.2, 0.2)),
+        ((0.99662423, 0.7520255, 0.00675154, 0.01446759), "yolo", (0.99324846, 0.744791705, 1.0, 0.759259295)),
+        ((0.9375, 0.510416, 0.1234375, 0.97638), "yolo", (0.87578125, 0.022226, 0.999218749, 0.998606)),
     ],
 )
 def test_convert_bbox_to_albumentations(bbox, source_format, expected):
@@ -87,7 +90,7 @@ def test_convert_bbox_to_albumentations(bbox, source_format, expected):
     converted_bbox = convert_bbox_to_albumentations(
         bbox, rows=image.shape[0], cols=image.shape[1], source_format=source_format
     )
-    assert converted_bbox == expected
+    assert np.all(np.isclose(converted_bbox, expected))
 
 
 @pytest.mark.parametrize(
@@ -97,8 +100,8 @@ def test_convert_bbox_to_albumentations(bbox, source_format, expected):
         ((0.2, 0.3, 0.6, 0.8, 99), "coco", (20, 30, 40, 50, 99)),
         ((0.2, 0.3, 0.6, 0.8), "pascal_voc", (20, 30, 60, 80)),
         ((0.2, 0.3, 0.6, 0.8, 99), "pascal_voc", (20, 30, 60, 80, 99)),
-        ((0.01, 0.06, 0.41, 0.56), "yolo", (0.2, 0.3, 0.4, 0.5)),
-        ((0.01, 0.06, 0.41, 0.56, 99), "yolo", (0.2, 0.3, 0.4, 0.5, 99)),
+        ((0.00, 0.05, 0.40, 0.55), "yolo", (0.2, 0.3, 0.4, 0.5)),
+        ((0.00, 0.05, 0.40, 0.55, 99), "yolo", (0.2, 0.3, 0.4, 0.5, 99)),
     ],
 )
 def test_convert_bbox_from_albumentations(bbox, target_format, expected):