Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Skip resizing if unnecessary #47

Open
wants to merge 2 commits into
base: main
Choose a base branch
from
Open

Skip resizing if unnecessary #47

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
4c1b191
Skip resizing if unnecessary
AurelienCoppee Feb 12, 2025
aaa58b6
Doesn't compute bbox if not used
AurelienCoppee Feb 13, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
39 changes: 21 additions & 18 deletions rtmlib/tools/object_detection/rtmdet.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,26 +38,29 @@ def preprocess(self, img: np.ndarray):

Returns:
tuple:
- resized_img (np.ndarray): Preprocessed image.
- center (np.ndarray): Center of image.
- scale (np.ndarray): Scale of image.
- padded_img (np.ndarray): Preprocessed image.
- ratio (float): Scale factor applied to the image.
"""
if len(img.shape) == 3:
padded_img = np.ones(
(self.model_input_size[0], self.model_input_size[1], 3),
dtype=np.uint8) * 114
if img.shape[:2] == tuple(self.model_input_size[:2]):
padded_img = img.copy()
ratio = 1.
else:
padded_img = np.ones(self.model_input_size, dtype=np.uint8) * 114

ratio = min(self.model_input_size[0] / img.shape[0],
self.model_input_size[1] / img.shape[1])
resized_img = cv2.resize(
img,
(int(img.shape[1] * ratio), int(img.shape[0] * ratio)),
interpolation=cv2.INTER_LINEAR,
).astype(np.uint8)
padded_shape = (int(img.shape[0] * ratio), int(img.shape[1] * ratio))
padded_img[:padded_shape[0], :padded_shape[1]] = resized_img
if len(img.shape) == 3:
padded_img = np.ones(
(self.model_input_size[0], self.model_input_size[1], 3),
dtype=np.uint8) * 114
else:
padded_img = np.ones(self.model_input_size, dtype=np.uint8) * 114

ratio = min(self.model_input_size[0] / img.shape[0],
self.model_input_size[1] / img.shape[1])
resized_img = cv2.resize(
img,
(int(img.shape[1] * ratio), int(img.shape[0] * ratio)),
interpolation=cv2.INTER_LINEAR,
).astype(np.uint8)
padded_shape = (int(img.shape[0] * ratio), int(img.shape[1] * ratio))
padded_img[:padded_shape[0], :padded_shape[1]] = resized_img

# normalize image
if self.mean is not None:
Expand Down
39 changes: 21 additions & 18 deletions rtmlib/tools/object_detection/yolox.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,26 +38,29 @@ def preprocess(self, img: np.ndarray):

Returns:
tuple:
- resized_img (np.ndarray): Preprocessed image.
- center (np.ndarray): Center of image.
- scale (np.ndarray): Scale of image.
- padded_img (np.ndarray): Preprocessed image.
- ratio (float): Scale factor applied to the image.
"""
if len(img.shape) == 3:
padded_img = np.ones(
(self.model_input_size[0], self.model_input_size[1], 3),
dtype=np.uint8) * 114
if img.shape[:2] == tuple(self.model_input_size[:2]):
padded_img = img.copy()
ratio = 1.
else:
padded_img = np.ones(self.model_input_size, dtype=np.uint8) * 114

ratio = min(self.model_input_size[0] / img.shape[0],
self.model_input_size[1] / img.shape[1])
resized_img = cv2.resize(
img,
(int(img.shape[1] * ratio), int(img.shape[0] * ratio)),
interpolation=cv2.INTER_LINEAR,
).astype(np.uint8)
padded_shape = (int(img.shape[0] * ratio), int(img.shape[1] * ratio))
padded_img[:padded_shape[0], :padded_shape[1]] = resized_img
if len(img.shape) == 3:
padded_img = np.ones(
(self.model_input_size[0], self.model_input_size[1], 3),
dtype=np.uint8) * 114
else:
padded_img = np.ones(self.model_input_size, dtype=np.uint8) * 114

ratio = min(self.model_input_size[0] / img.shape[0],
self.model_input_size[1] / img.shape[1])
resized_img = cv2.resize(
img,
(int(img.shape[1] * ratio), int(img.shape[0] * ratio)),
interpolation=cv2.INTER_LINEAR,
).astype(np.uint8)
padded_shape = (int(img.shape[0] * ratio), int(img.shape[1] * ratio))
padded_img[:padded_shape[0], :padded_shape[1]] = resized_img

return padded_img, ratio

Expand Down
39 changes: 21 additions & 18 deletions rtmlib/tools/pose_estimation/rtmo.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -48,26 +48,29 @@ def preprocess(self, img: np.ndarray):

Returns:
tuple:
- resized_img (np.ndarray): Preprocessed image.
- center (np.ndarray): Center of image.
- scale (np.ndarray): Scale of image.
- padded_img (np.ndarray): Preprocessed image.
- ratio (float): Scale factor applied to the image.
"""
if len(img.shape) == 3:
padded_img = np.ones(
(self.model_input_size[0], self.model_input_size[1], 3),
dtype=np.uint8) * 114
if img.shape[:2] == tuple(self.model_input_size[:2]):
padded_img = img.copy()
ratio = 1.
else:
padded_img = np.ones(self.model_input_size, dtype=np.uint8) * 114

ratio = min(self.model_input_size[0] / img.shape[0],
self.model_input_size[1] / img.shape[1])
resized_img = cv2.resize(
img,
(int(img.shape[1] * ratio), int(img.shape[0] * ratio)),
interpolation=cv2.INTER_LINEAR,
).astype(np.uint8)
padded_shape = (int(img.shape[0] * ratio), int(img.shape[1] * ratio))
padded_img[:padded_shape[0], :padded_shape[1]] = resized_img
if len(img.shape) == 3:
padded_img = np.ones(
(self.model_input_size[0], self.model_input_size[1], 3),
dtype=np.uint8) * 114
else:
padded_img = np.ones(self.model_input_size, dtype=np.uint8) * 114

ratio = min(self.model_input_size[0] / img.shape[0],
self.model_input_size[1] / img.shape[1])
resized_img = cv2.resize(
img,
(int(img.shape[1] * ratio), int(img.shape[0] * ratio)),
interpolation=cv2.INTER_LINEAR,
).astype(np.uint8)
padded_shape = (int(img.shape[0] * ratio), int(img.shape[1] * ratio))
padded_img[:padded_shape[0], :padded_shape[1]] = resized_img

# normalize image
if self.mean is not None:
Expand Down
2 changes: 1 addition & 1 deletion rtmlib/tools/solution/pose_tracker.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -185,7 +185,7 @@ def __call__(self, image: np.ndarray):
keypoints, scores = self.pose_model(image)


if not self.tracking:
if not self.tracking and self.det_frequency != 1:
# without tracking
bboxes_current_frame = []
for kpts in keypoints:
Expand Down