search

javiribera / locating-objects-without-bboxes

PyTorch code for "Locating objects without bounding boxes" - Loss function and trained models
Other
249 stars 52 forks source link

Adapting your loss to segmentation #27

Open TheoEst opened 4 years ago

TheoEst commented 4 years ago

Hello, Thanks for this nice paper and the code. Of my understanding, in you code (WeightedHausdorffDistance class), the ground truth gt is a list of point corresponding to different objects. Is it possible to have gt a list of point from the same object ?

I am wondering of using your loss for a segmentation task in medical imaging to replace (or in parallel of) Dice Loss. So we have the ground truth segmentation of one object.

Best Regards, Théo

RSKothari commented 4 years ago

@TheoEst I can confirm that this loss works well in parallel with segmentation loss. We are in the process of publishing our work. Will update this response with a link to our Arxiv paper.

Syzygianinfern0 commented 3 years ago

@TheoEst I can confirm that this loss works well in parallel with segmentation loss. We are in the process of publishing our work. Will update this response with a link to our Arxiv paper.

@RSKothari Can you please provide the link to your paper.

RSKothari commented 3 years ago

@Syzygianinfern0 We decided to not use this loss as it did not add any advantage (in terms of segmentation metrics). You can find the project here. Official paper will be published at IEEE-VR.