infrontofme / uwgan_uie Goto Github PK
View Code? Open in Web Editor NEWSource code for UWGAN: Underwater GAN for Real-world Underwater Color Restoration and Dehazing
Home Page: https://arxiv.org/abs/1912.10269
License: MIT License
Source code for UWGAN: Underwater GAN for Real-world Underwater Color Restoration and Dehazing
Home Page: https://arxiv.org/abs/1912.10269
License: MIT License
Hello, Can you please share the checkpoint folder with us? which could save lots of time for people. It will be greatly appreciated.
Thanks.
我使用了作者提供的训练模型,但是复现出的结果还是很差,想知道在test_one_image.py中的gt_image_path是什么?在测试的时候不是应该输入测试图片,然后加载模型就可以了吗?
谢谢啦
@infrontofme Can you please share the checkpoints
作者你好,
我试着训练了你的model,但在result_color中并未产生图像,
请问可能是甚麽原因呢
Hi, thank you for sharing the project. Can you share the link of trained model?Because of the limit of space of disk, I can't train the model.
hello,i want to see what's journal published your paper?
thank you!
我使用了作者提供的数据集训练UNet,参数的设定和论文提供的参数设定完全相同,为什么复现不了UNet的修复和去雾结果?生成的图片基本什么也看不到,这是什么原因?求大佬帮助
Hello, I am interested in trying your method on my own underwater images.
However, downloading full NYU image sets requires for V1 90GB (Raw images) and for v2 it's 428GB for V2. I simply don't have enough space or time to wait to download all this.
So, I was wondering if you could upload pre-trained UWGAN/U-Net models instead that one could simply download and try on real life underwater images? It would be a great help.
大佬好,非常感谢你的工作分享,确实很赞(大拇哥)!
我自己跑了一下code,发现模拟雾化的3个参数(对应r, g, b三个通道)并未参与到训练中(一直保持0.75),请问这样的话当遇到有雾化效果的水下图像的时候,会影响增强效果吧?
你好,我在执行源代码上会出现版本不相容的状况。
想请教可以使用的环境版本是多少,谢谢
cuda跟cuDNN也拜託了。
哥们留个联系方式吗?我fsdet一直编译不过,一定要装cuda92的吗?清华原里面找不到了
It will be helpful if it contains a sample test image and it's usage.So it will help to quickly run and test the model weather is working or not before testing it with custom data also it will be helpful to understand the basic working
It seems that UWGAN generated properly fake water images, it also seems that UNet trained properly on those images as I can get back original image from fake when I run test. However, if I try to feed test with actual water image I get out garbage. I am including sample fake image generated by UWGAN. The same image reconstructed by UNet (seems properly) and what I get when I try to feed Unet one of Type1 water images.
(fake water image generated by UWGAN)
(the same image fed through test loop of UNet (after training it of course))
(sample Type1 water image after going through UNet)
Any idea, what might be going wrong?
Any help would be appreciated.
Thx for uploading to GDrive.
I have unpacked everything and before I can try training there is one more issue left: There are 2 subfolders for water_images (Type1 and type2) while there is only one for air_images.
Should Type1 and Type2 water images be combined into one folder and Type1 files renamed somehow to get correct pairing with air images?
大佬好,有个问题请教,当我train uNet的时候,出现train loss nan, cost 0.000的情况,trainA和trainB是利用UWGAN生成的图像数据,超参采用的都是默认值,希望能够帮助解答,谢谢
I trained the model with UWGAN, but results_color1 folder didn't generate pictures
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.