search

dbolya / tide

A General Toolbox for Identifying Object Detection Errors
https://dbolya.github.io/tide
MIT License
702 stars 115 forks source link

Quality of Mask for evaluation #12

Closed abhigoku10 closed 3 years ago

abhigoku10 commented 3 years ago

@dbolya @hyperparameters can we also some parameters to define the quality of the mask , eg: how much difference is there wrt to groundtruth the edges basically

dbolya commented 3 years ago

I'm not sure what you mean by this? IoU already captures very similar information, so Mask localization error should already capture mask quality.

abhigoku10 commented 3 years ago

@dbolya If the prediction mask and gt mask has IOU overlap >0.5 and yet the edges / detail contours of the object are not correct how to estimate the difference between them or basically the difference in the detailed shape of the masks