ReaFly / ACSNet

MICCAI 2020 : Adaptive Context Selection for Polyp Segmentation (Pytorch implementation).
54 stars 4 forks source link

How should I get the segmentation result picture? #4

Open tjpustoy opened 2 years ago

tjpustoy commented 2 years ago

I fllowed this work, but I noticed that there is a comparison of the segmentation results in the paper. After the test is completed, no result picture is generated. Is this part of the code missing? Hope to update the code that saves the result picture, thank you, this is really important to me.

ReaFly commented 2 years ago

Thanks for your attention. You can simply set the threshold as 0.5 to convert the probability map to the binary segmentation result and then save it.

WangZhi-wz commented 2 years ago

我完成了这项工作,但我注意到论文中对分割结果进行了比较。测试完成后,没有生成结果图片。这部分代码是否丢失? 希望更新保存结果图片的代码,谢谢,这对我来说真的很重要。

请问你完成这个工作了么,我也在尝试生成结果图片,但是我是初学者,并不是太理解,请问可以粘贴一下生成图片的代码么

tjpustoy commented 2 years ago

You can simply set the threshold as 0.5 to convert the probability map to the binary segmentation result and then save it.