ddfnet
ddfnet copied to clipboard
The official implementation of the CVPR2021 paper: Decoupled Dynamic Filter Networks
对于spatial和channel滤波器是像素相乘完在与特征图谱做卷积乘法,还是通道滤波器先与输入特征图谱先完成卷积乘法,然后在与spatial滤波器相乘
Thank you for the open-source code. When using the module 'DDFPack', what is the difference between the 'add' and 'mul' combination mode?
The kernel size can not be larger than 13x13? I set the kernel size to 15x15 and encounter the error "CUDA error: an illegal memory access was encountered". But everything...
Thank you for the open-source code. In your code, the kernel of ddf is set to square (i.e. 3x3), if I want to set the kernel to a normal rectangle...
Thanks for your excellent work and I met some problems when I train your model following your instruction. 1. You claimed in your paper that you were using batch size...
DDFPack and DDFUpPack currently assume that the in_channels and out_channels are the same. Can you make short changes to the modules to allow them to work for different in_channels and...
Looks like DDFUnPack is incomplete. Would you please also release the DDF-Up-Net example codes? Thanks,
Hello,I met this question,can you help me? python grad_check.py Traceback (most recent call last): File "grad_check.py", line 8, in from ddf import ddf File "/media/omnisky/34B22D6336AC1687/JSS/ddfnet-main/ddf/__init__.py", line 1, in from .ddf...
from . import ddf_mul_ext, ddf_mul_faster_ext, ddf_add_ext, ddf_add_faster_ext Can anyone please help me to know what does this dot after from word in the above line indicates (. ) means in...
sir, can you please let me know, why have you use backward compatible code also in the implementation. I am talking about the class defined below.. @staticmethod def backward(ctx, grad_output):...