yjn870/arcnn-pytorch

PyTorch implementation of Deep Convolution Networks for Compression Artifacts Reduction (ICCV 2015)

arcnn-pytorch's Issues

Excuse me, where can I get the dataset that you used for training?

why??my result is not RGB

dataset problem

I want to know where can I find the dataset or weights path if it is possible ?

Training goal

Thanks for the work. A question is that I do not understand why you train on jpeg-corrupted images while in the SRCNN paper it says that they downscale original images by upscale factor (for example, 3): "To synthesize the low-resolution samples {Yi}, we blur a sub-image by a Gaussian kernel, sub-sample it by the upscaling factor, and upscale it by the same factor via bicubic interpolation.". And I think it is the proper goal for the singe image super-resolution restoration task.

cuda error:unspecified launch failure

when I run main.py, I got this error message.

Traceback (most recent call last):
    File "main.py'', line 66, in <module>
         inputs = inputs.to(device)
RuntimeError: CUDA error: unspecified launch failure