search

layumi / 2016_super_resolution

ICCV2015 Image Super-Resolution Using Deep Convolutional Networks
MIT License
86 stars 43 forks source link

Have a question #1

Open jhcho90 opened 8 years ago

jhcho90 commented 8 years ago

Hi I'm a student who is interested in Computer Vision.

Nowadays, I especially study Super-resolution.

So I read paper and chance to download your code.

But I have some problems.

I have a question

1) gpu_compile ( I use cuda7.0, matlab R2014, cudnn-7.0-win-x64-v4.0) is okay but when I using 'testRnet_result' is not working that it can't find mex-file vl_nnconv.mexw64 so I replace the other vl_nnconv.mexw64

2) I wanna train myself when compile train_SRnet line50 is not working because vl_nnconv error

Is it possible to get any comment? I really anticipate. Thanks.

MarkYF commented 8 years ago

glad to have you letter.

as far as I remeber I didn't post my codes onto Github. maybe you can post me the copy.

about vl_nnconv.mexw64, I think you should work on windows instead of mac or linux, cause mexw64 is a kind of code of win.

If you already got the windows, I don't konw what's wrong with your code.

Glad to discuss with you.

2016-08-26 17:08 GMT+08:00, rehoon notifications@github.com:

Hi I'm a student who is interested in Computer Vision.

Nowadays, I especially study Super-resolution.

So I read paper and chance to download your code.

But I have some problems.

I have a question

1) gpu_compile ( I use cuda7.0, matlab R2014, cudnn-7.0-win-x64-v4.0) is okay but when I using 'testRnet_result' is not working that it can't find mex-file vl_nnconv.mexw64 so I replace the other vl_nnconv.mexw64

2) I wanna train myself when compile train_SRnet line50 is not working because vl_nnconv error

Is it possible to get any comment? I really anticipate. Thanks.

You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/layumi/2016_super_resolution/issues/1

Best Regards WangYunfei（王云飞）

Information Engineering College Capital Normal University Beijing 100048, China

E-mail: mark.wyf@gmail.com

layumi commented 8 years ago

@rehoon my super resolution code is trained on Ubuntu 14.04 You may refer to http://www.vlfeat.org/matconvnet/install/ to install matconvnet And then you may try to run gpu_compile.m

Addhi86 commented 8 years ago

@layumi is this is the actual paper, Deep Networks for Image Super-Resolution with Sparse Prior? Also, can you please change the link of the dataset. As when i tried to download, its shows that i must have to download Baidu and make an account and I don't understand Chinese.

Thanks

layumi commented 8 years ago

@Addhi86 I follow the setting in the paper Image Super-Resolution Using Deep Convolutional Networks C.Dong (https://arxiv.org/pdf/1501.00092.pdf)

I will update README.md to show how to download from Baidu Driver.

Addhi86 commented 8 years ago

@layumi Thanks for replying me back and updating the README. As you have mentioned when I click the download it shows me this thing. I have attached the picture capture

layumi commented 8 years ago

@Addhi86 Would you click into the folder and download three files without install the software? Baidu is so annoying. Baidu seems to force Windows user to install their software. You may try to download them on a Mac? I can download it without installation on Mac.

Addhi86 commented 8 years ago

@layumi No I cant. can you share these files on my email? ednan4@gmail.com

layumi commented 8 years ago

@Addhi86 these files are 6 GB... You may install their free software to download and uninstall it after download. lol In China, uploading files to Google Driver is too slow. I failed.