Fix xyxyxyxy2xywhr for Numpy inputs (#13273)

ultralytics/utils/ops.py

@@ -518,59 +518,58 @@ def ltwh2xywh(x):
     return y
 
 
-def xyxyxyxy2xywhr(corners):
+def xyxyxyxy2xywhr(x):
     """
     Convert batched Oriented Bounding Boxes (OBB) from [xy1, xy2, xy3, xy4] to [xywh, rotation]. Rotation values are
     expected in degrees from 0 to 90.
 
     Args:
-        corners (numpy.ndarray | torch.Tensor): Input corners of shape (n, 8).
+        x (numpy.ndarray | torch.Tensor): Input box corners [xy1, xy2, xy3, xy4] of shape (n, 8).
 
     Returns:
         (numpy.ndarray | torch.Tensor): Converted data in [cx, cy, w, h, rotation] format of shape (n, 5).
     """
-    is_torch = isinstance(corners, torch.Tensor)
+    is_torch = isinstance(x, torch.Tensor)
-    points = corners.cpu().numpy() if is_torch else corners
+    points = x.cpu().numpy() if is_torch else x
-    points = points.reshape(len(corners), -1, 2)
+    points = points.reshape(len(x), -1, 2)
     rboxes = []
     for pts in points:
         # NOTE: Use cv2.minAreaRect to get accurate xywhr,
         # especially some objects are cut off by augmentations in dataloader.
-        (x, y), (w, h), angle = cv2.minAreaRect(pts)
+        (cx, cy), (w, h), angle = cv2.minAreaRect(pts)
-        rboxes.append([x, y, w, h, angle / 180 * np.pi])
+        rboxes.append([cx, cy, w, h, angle / 180 * np.pi])
-    return (
+    return torch.tensor(rboxes, device=x.device, dtype=x.dtype) if is_torch else np.asarray(rboxes)
-        torch.tensor(rboxes, device=corners.device, dtype=corners.dtype)
-        if is_torch
-        else np.asarray(rboxes, dtype=points.dtype)
-    )  # rboxes
 
 
-def xywhr2xyxyxyxy(rboxes):
+def xywhr2xyxyxyxy(x):
     """
     Convert batched Oriented Bounding Boxes (OBB) from [xywh, rotation] to [xy1, xy2, xy3, xy4]. Rotation values should
     be in degrees from 0 to 90.
 
     Args:
-        rboxes (numpy.ndarray | torch.Tensor): Boxes in [cx, cy, w, h, rotation] format of shape (n, 5) or (b, n, 5).
+        x (numpy.ndarray | torch.Tensor): Boxes in [cx, cy, w, h, rotation] format of shape (n, 5) or (b, n, 5).
 
     Returns:
         (numpy.ndarray | torch.Tensor): Converted corner points of shape (n, 4, 2) or (b, n, 4, 2).
     """
-    is_numpy = isinstance(rboxes, np.ndarray)
+    cos, sin, cat, stack = (
-    cos, sin = (np.cos, np.sin) if is_numpy else (torch.cos, torch.sin)
+        (torch.cos, torch.sin, torch.cat, torch.stack)
+        if isinstance(x, torch.Tensor)
+        else (np.cos, np.sin, np.concatenate, np.stack)
+    )
 
-    ctr = rboxes[..., :2]
+    ctr = x[..., :2]
-    w, h, angle = (rboxes[..., i : i + 1] for i in range(2, 5))
+    w, h, angle = (x[..., i : i + 1] for i in range(2, 5))
     cos_value, sin_value = cos(angle), sin(angle)
     vec1 = [w / 2 * cos_value, w / 2 * sin_value]
     vec2 = [-h / 2 * sin_value, h / 2 * cos_value]
-    vec1 = np.concatenate(vec1, axis=-1) if is_numpy else torch.cat(vec1, dim=-1)
+    vec1 = cat(vec1, -1)
-    vec2 = np.concatenate(vec2, axis=-1) if is_numpy else torch.cat(vec2, dim=-1)
+    vec2 = cat(vec2, -1)
     pt1 = ctr + vec1 + vec2
     pt2 = ctr + vec1 - vec2
     pt3 = ctr - vec1 - vec2
     pt4 = ctr - vec1 + vec2
-    return np.stack([pt1, pt2, pt3, pt4], axis=-2) if is_numpy else torch.stack([pt1, pt2, pt3, pt4], dim=-2)
+    return stack([pt1, pt2, pt3, pt4], -2)
 
 
 def ltwh2xyxy(x):
@@ -785,7 +784,7 @@ def regularize_rboxes(rboxes):
     Regularize rotated boxes in range [0, pi/2].
 
     Args:
-        rboxes (torch.Tensor): (N, 5), xywhr.
+        rboxes (torch.Tensor): Input boxes of shape(N, 5) in xywhr format.
 
     Returns:
         (torch.Tensor): The regularized boxes.