SRMD
SRMD copied to clipboard
cszn
Learning a Single Convolutional Super-Resolution Network for Multiple Degradations (CVPR, 2018) (Matlab)
Thanks for your amazing work! But I have a question that does " synthesize an LR image with spatially variant blur kernels and noise levels" in 4.4 mean one image...
hello, I have a question about the experiment on bicubic degradation. in this experiment, how to get the blur kernel to form degradation maps.Thank you for your extremely busy schedule...
Hello,I can not understand one sentence in your paper,'When the training error keeps unchanged in five sequential epochs, we merge the parameters of each batch normalization into the adjacent convolution...
I have some confusion about the generation of LR images with different spatially variant degradations, can you provide more details? Thank you very much!
I find the file 'net.meta.P' in your codes, whose size is 15x255, could you tell me how to make it? And it seems that your model uses subpixel with upsample_factor=0.5...
In your paper: We set the rotation angle range to [0,π]. For the scaling of eigenvalues, it is set from 0.5 to 6, 8 and 10 for scale factors 2,...
Hi, thank you so much for the source code. Besides, could you please provide the resulting image of the degradation model?
>> Demo_real_application_denoising_and_deblurring_color -------------------------------------------- ----frog-----Super-Resolution----- -------------------------------------------- Undefined function or variable 'vl_argparse'. Error in vl_nnSubP (line 5) opts = vl_argparse(opts, varargin, 'nonrecursive') ; Error in vl_srmd_matlab (line 36) res(i+1).x = vl_nnSubP(res(i).x, [],'scale',l.scale);...
The training is normal. The following problems are encountered in the testing process. wrong with vl_nnbnorm vl::impl::dispatch_cudnn::operator(): BatchNormForwardWithMomentCudnn::operator(): cuDNN error [cudnn: "d:\\matlab\\r2018b\\bin\\matconvnet-1.0-beta25\\matlab\\src\\bits\\nnbnorm_cudnn.cu":155 (CUDNN_STATUS_BAD_PARAM)]
