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.

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

No keypoints with sift = error #114

Open
ExtReMLapin opened this issue Feb 15, 2024 · 0 comments
Open

No keypoints with sift = error #114

ExtReMLapin opened this issue Feb 15, 2024 · 0 comments

Comments

@ExtReMLapin
Copy link

ExtReMLapin commented Feb 15, 2024

Hello, only when using sift I get this error on some images:


Traceback (most recent call last):
  File "server.py", line 510, in post
    scores = classify_from_color_layers(colors, just_do_class=force_class_param)
  File "server.py", line 219, in classify_from_color_layers
    image_features = get_features(image_np)
  File "server.py", line 84, in get_features
    feats0 = extractor.extract(image0.to(device))
  File "venv\lib\site-packages\torch\utils\_contextlib.py", line 115, in decorate_context
    return func(*args, **kwargs)
  File "venv\lib\site-packages\lightglue\utils.py", line 144, in extract
    feats = self.forward({"image": img})
  File "venv\lib\site-packages\lightglue\sift.py", line 211, in forward
    p = self.extract_single_image(img)
  File "venv\lib\site-packages\lightglue\sift.py", line 178, in extract_single_image
    keep = filter_dog_point(
  File "venv\lib\site-packages\lightglue\sift.py", line 25, in filter_dog_point
    np.maximum.at(buffer, tuple(ij), s)
ValueError: array is not broadcastable to correct shape

Tried to fix it myself but I failed, pred["keypoints"] len is 0

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant