3D medical image process

[shengzhang90]

Sorry to bother you, Xing, and thanks for your contribution to the repository of https://github.com/XinyiYing/D3Dnet, Could you please modify the code from 2N -> 3N, which can estimate the offsets of the three directions (time, height, width), not only the two offsets of the two directions (height, width) with a 3D input? For I want to apply it in the 3D medical image process. If applicable, I would cite your papers and publicize it positively. Thanks a lot.

[XinyiYing]

You can choose any dimension you want to deform by setting parameter dimension. See code/dcn/test.py for more details.
Example codes of deform among three dimension:
dcn = DeformConvPack_d(inC, outC, kernel_size=[kT, kH, kW], stride=[sT, sH, sW],padding=[pT, pH, pW],dimension='THW').cuda()

[shengzhang90]

I have viewed all the issues in the repository. Now, I can not compile it with the following issue:
subprocess.CalledProcessError: Command '['ninja', '-v']' returned non-zero exit status 1.

Even though I checked the environment and tried out the installation of ninja application the whole evening yesterday carefully.

Thanks a lot for a pretty and kind girl programmer of you.

[XinyiYing]

We recommend users to use ubuntu for compiling. Many users who use win for compiling always report a variety of environmental bugs, including the ninja error.