Added keypoints support to RandomSizedBBoxSafeCrop.

README.md

@@ -182,7 +182,7 @@ Spatial-level transforms will simultaneously change both an input image as well
 | [RandomResizedCrop](https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.RandomResizedCrop) | ✓ | ✓ | ✓ | ✓ |
 | [RandomRotate90](https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.RandomRotate90) | ✓ | ✓ | ✓ | ✓ |
 | [RandomScale](https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.RandomScale) | ✓ | ✓ | ✓ | ✓ |
-| [RandomSizedBBoxSafeCrop](https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.RandomSizedBBoxSafeCrop) | ✓ | ✓ | ✓ |   |
+| [RandomSizedBBoxSafeCrop](https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.RandomSizedBBoxSafeCrop) | ✓ | ✓ | ✓ | ✓ |
 | [RandomSizedCrop](https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.RandomSizedCrop) | ✓ | ✓ | ✓ | ✓ |
 | [Resize](https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.Resize) | ✓ | ✓ | ✓ | ✓ |
 | [Rotate](https://albumentations.ai/docs/api_reference/augmentations/transforms/#albumentations.augmentations.transforms.Rotate) | ✓ | ✓ | ✓ | ✓ |

albumentations/augmentations/functional.py

@@ -1502,16 +1502,16 @@ def bbox_flip(bbox, d, rows, cols):
 
 
 def crop_bbox_by_coords(bbox, crop_coords, crop_height, crop_width, rows, cols):
- """Crop a bounding box using the provided coordinates of bottom-left and top-right corners in pixels and the
- required height and width of the crop.
+ """Crop a bounding box using the provided coordinates `crop_coords` of bottom-left and top-right corners in pixels
+  and scale it to fit cropped area being scaled to `(crop_height, crop_width)` size.
 
  Args:
  bbox (tuple): A cropped box `(x_min, y_min, x_max, y_max)`.
- crop_coords (tuple): Crop coordinates `(x1, y1, x2, y2)`.
- crop_height (int):
- crop_width (int):
- rows (int): Image rows.
- cols (int): Image cols.
+ crop_coords (tuple): Crop area coordinates in an original image `(x1, y1, x2, y2)`.
+ crop_height (int): Target height of cropped area after scaling.
+ crop_width (int): Target width of cropped area after scaling.
+ rows (int): Original image height.
+ cols (int): Original image width.
 
  Returns:
  tuple: A cropped bounding box `(x_min, y_min, x_max, y_max)`.
@@ -1771,17 +1771,12 @@ def keypoint_scale(keypoint, scale_x, scale_y):
  return x * scale_x, y * scale_y, angle, scale * max(scale_x, scale_y)
 
 
-def crop_keypoint_by_coords(keypoint, crop_coords, crop_height, crop_width, rows, cols): # skipcq: PYL-W0613
- """Crop a keypoint using the provided coordinates of bottom-left and top-right corners in pixels and the
- required height and width of the crop.
+def crop_keypoint_by_coords(keypoint, crop_coords): # skipcq: PYL-W0613
+ """Crop a keypoint using the provided coordinates `crop_coords` of bottom-left and top-right corners in pixels
 
  Args:
  keypoint (tuple): A keypoint `(x, y, angle, scale)`.
- crop_coords (tuple): Crop box coords `(x1, x2, y1, y2)`.
- crop height (int): Crop height.
- crop_width (int): Crop width.
- rows (int): Image height.
- cols (int): Image width.
+ crop_coords (tuple): Crop area coords `(x1, x2, y1, y2)`. Actually only `(x1,y1)` matters.
 
  Returns:
  A keypoint `(x, y, angle, scale)`.
@@ -1809,7 +1804,7 @@ def keypoint_random_crop(keypoint, crop_height, crop_width, h_start, w_start, ro
 
  """
  crop_coords = get_random_crop_coords(rows, cols, crop_height, crop_width, h_start, w_start)
- return crop_keypoint_by_coords(keypoint, crop_coords, crop_height, crop_width, rows, cols)
+ return crop_keypoint_by_coords(keypoint, crop_coords)
 
 
 def keypoint_center_crop(keypoint, crop_height, crop_width, rows, cols):
@@ -1829,7 +1824,7 @@ def keypoint_center_crop(keypoint, crop_height, crop_width, rows, cols):
 
  """
  crop_coords = get_center_crop_coords(rows, cols, crop_height, crop_width)
- return crop_keypoint_by_coords(keypoint, crop_coords, crop_height, crop_width, rows, cols)
+ return crop_keypoint_by_coords(keypoint, crop_coords)
 
 
 def py3round(number):

albumentations/augmentations/transforms.py

@@ -198,14 +198,7 @@ def apply_to_bbox(self, bbox, **params):
  return F.bbox_crop(bbox, x_min=self.x_min, y_min=self.y_min, x_max=self.x_max, y_max=self.y_max, **params)
 
  def apply_to_keypoint(self, keypoint, **params):
- return F.crop_keypoint_by_coords(
- keypoint,
- crop_coords=(self.x_min, self.y_min, self.x_max, self.y_max),
- crop_height=self.y_max - self.y_min,
- crop_width=self.x_max - self.x_min,
- rows=params["rows"],
- cols=params["cols"],
- )
+ return F.crop_keypoint_by_coords(keypoint, crop_coords=(self.x_min, self.y_min, self.x_max, self.y_max))
 
  def get_transform_init_args_names(self):
  return ("x_min", "y_min", "x_max", "y_max")
@@ -778,14 +771,7 @@ def apply_to_bbox(self, bbox, x_min=0, x_max=0, y_min=0, y_max=0, **params):
  return F.bbox_crop(bbox, y_max - y_min, x_max - x_min, h_start, w_start, **params)
 
  def apply_to_keypoint(self, keypoint, x_min=0, x_max=0, y_min=0, y_max=0, **params):
- return F.crop_keypoint_by_coords(
- keypoint,
- crop_coords=(x_min, y_min, x_max, y_max),
- crop_height=y_max - y_min,
- crop_width=x_max - x_min,
- rows=params["rows"],
- cols=params["cols"],
- )
+ return F.crop_keypoint_by_coords(keypoint, crop_coords=(x_min, y_min, x_max, y_max))
 
  @property
  def targets_as_params(self):
@@ -953,6 +939,8 @@ def get_transform_init_args_names(self):
 
 class RandomSizedBBoxSafeCrop(DualTransform):
  """Crop a random part of the input and rescale it to some size without loss of bboxes.
+ Keypoints are processed but does not affect selection of crop region. So if you need to crop preserving
+ keypoints you have to pass fictive bbox enclosing keypoints.
 
  Args:
  height (int): height after crop and resize.
@@ -964,7 +952,7 @@ class RandomSizedBBoxSafeCrop(DualTransform):
  p (float): probability of applying the transform. Default: 1.
 
  Targets:
- image, mask, bboxes
+ image, mask, bboxes, keypoints
 
  Image types:
  uint8, float32
@@ -1009,6 +997,13 @@ def get_params_dependent_on_targets(self, params):
  def apply_to_bbox(self, bbox, crop_height=0, crop_width=0, h_start=0, w_start=0, rows=0, cols=0, **params):
  return F.bbox_random_crop(bbox, crop_height, crop_width, h_start, w_start, rows, cols)
 
+ def apply_to_keypoint(self, keypoint, crop_height=0, crop_width=0, h_start=0, w_start=0, rows=0, cols=0, **params):
+ return F.keypoint_scale(
+ F.keypoint_random_crop(keypoint, crop_height, crop_width, h_start, w_start, rows, cols),
+ self.width / crop_width,
+ self.height / crop_height,
+ )
+
  @property
  def targets_as_params(self):
  return ["image", "bboxes"]
@@ -1058,14 +1053,7 @@ def apply_to_bbox(self, bbox, x_min=0, x_max=0, y_min=0, y_max=0, **params):
  )
 
  def apply_to_keypoint(self, keypoint, x_min=0, x_max=0, y_min=0, y_max=0, **params):
- return F.crop_keypoint_by_coords(
- keypoint,
- crop_coords=[x_min, y_min, x_max, y_max],
- crop_height=y_max - y_min,
- crop_width=x_max - x_min,
- rows=params["rows"],
- cols=params["cols"],
- )
+ return F.crop_keypoint_by_coords(keypoint, crop_coords=[x_min, y_min, x_max, y_max])
 
  @property
  def targets_as_params(self):