saskra

I will try. It might be based on this: https://github.com/WillBrennan/SemanticSegmentation/blob/7646661aa0e05f5c5b087f2aeda3faff42c34e00/semantic_segmentation/dataset.py#L103

I am also stuck with this error message - how did you get rid of it? And how to prevent it from coming back with the next update?

Same problem here!

I can confirm that blurred image detection gives way too many false positives, while the actual blurred ones are not even among the top ranks. The other issue types work...

This is a dataset of >7,500 smartphone photos of various resolutions, of which >500 were detected as blurry, even though it should be a low double-digit number. Unfortunately, I cannot...