search

cheng-01037 / Self-supervised-Fewshot-Medical-Image-Segmentation

[ECCV'20] Self-supervision with Superpixels: Training Few-shot Medical Image Segmentation without Annotation (code&data-processing pipeline)
MIT License
397 stars 50 forks source link

Request Code Help #40

Open WangBingJian233 opened 1 year ago

WangBingJian233 commented 1 year ago

Hello author, thank you for your outstanding work! Can you please publish the source code for your Extended Article "Self-Supervised Learning for Few-Shot Medical Image Segmentation,"(doi: 10.1109/TMI.2022.3150682) that was accepted on TMI in 2022? I would like to know how you set up the 5-shot during your testing

sadimanna commented 1 year ago

In relation to the above question, I have tried using sobel from kornia.filter but I haven't been able to replicate the results reported in the extended version. Any information about how the boundary loss is implemented would be useful.

cheng-01037 commented 1 year ago

They are based on a quick implementation with pytorch + fixed kernel. Will update with you soon.

sadimanna commented 7 months ago

@cheng-01037 Did you use a fixed convolutional kernel? A little nudge in the right direction will be greatly appreciated.