Siqi Liu

For the same architecture we did not have the numbers for deconv layers. Generally deconv layers tend to give the high frequency artifacts in the early epochs. So they were...

Thanks for your interests. You are right. The 2D network code here is slightly different to what we used in the paper. It is an easy-to-read implementation adapted from https://github.com/ycszen/pytorch-segmentation/blob/master/gcn.py...

Not sure if I understood your question. True positive here means true positive markers that can be catched by a lesion finding annotated by the radiologist (3D bounding box). There...

+1 Also due to the absolute importing, it now only works if you copy the human_controller.py to the project root

@argunsah Hi, thanks for your interests and reporting this issue. I can confirm the issue and I am looking into it. I have fixed it by adding the pad and...

also the `--soma` flag has been deprecated (I just updated the readme). Now if you add `--quality` the tracing function will use the SRET algorithm to reconstruct the soma contour....

The second issue is caused by the `pyglet` package we used for the simple rendering function. I currently cannot reproduce this on my side, I guess it was caused by...

If you have `--save-soma` and `--quality`, the reconstructed soma mask is saved to *.tif.r2.soma.tif in the same folder as your input tif if reconstructed correctly. For image filtering, there is...

Thanks for your interests. Not sure if I understood your question correctly. Do you mean to rasterize the vtk / swc result lines in a 3D numpy array? so that...

Thanks for reporting. It seems like a bug caused by pip10. I will try to make a quick fix and let you know. On Wed, Sep 11, 2019, 12:32 PM...