xindongzhang / elan Goto Github PK
View Code? Open in Web Editor NEW[ECCV2022] Efficient Long-Range Attention Network for Image Super-resolution
License: Apache License 2.0
[ECCV2022] Efficient Long-Range Attention Network for Image Super-resolution
License: Apache License 2.0
作者您好,您什么能够开源代码!
self.weight[0g:1g, 0, 1, 2] = 1.0 ## left
self.weight[1g:2g, 0, 1, 0] = 1.0 ## right
self.weight[2g:3g, 0, 2, 1] = 1.0 ## up
self.weight[3g:4g, 0, 0, 1] = 1.0 ## down
self.weight[4*g:, 0, 1, 1] = 1.0 ## identity
i think [1,2] is down,[1,0] is up,[2,1] is right,[0,1] is left.
Hello,
I am running the code and in second epoch, it says the in and out tensor size issue. The in is 3,256,256 and out is 3,248,258. I tried to use resize function. I have used the following configuration:
batch_size: 8
data_repeat: 80
data_augment: 1
epochs: 1000
lr: 0.00025
Please guide!
计算复杂度的代码有么? 我们算出来的和论文的不同,希望可以给个参考。
输入
input = torch.randn(1, 3, 320, 180)
报错如题 。
请问怎么修改
网络配置:
When training the ELAN model for SR ×3 and ×4, do we use the pre-trained ×2 network to initialize the model parameters?
Thank for your work and congratulate accept for ECCV2022.
I read your paper and had a question about other model metric.
In Table1 in your paper, I think that LatticeNet metric reports (PSNR/SSIM) are wrong.
Your report about that and the reports in both LatticeNet and SwinIR paper are different.
So I guess if you did experiments with LatticeNet by yourself, or if it is just mistake.
Thanks for your work one more.
I am reproducing your scores in your paper, and all looks well. Your framework is the neatest one I have seen. Anyway, thank you.
Hello,
The paper is very interesting to me, since SwinIR suffers from high memory consumption and slow convergence. I recently have two questions about the proposed framework.
Firstly, two consecutive GMSAs can share the attention maps, while the shifted window is applied to partition neighboring pixels together, which should derive different attention patterns. How is it addressed or is interleaved sharing mechanism adopted?
Secondly, the results in Table 3 show the reduction of FLOPs and Latency by using the shifted mechanism. How could this method reduce the computational footprint? Is it solely due to the removal of the masking and relative positional encoding used in SwinIR?
Finally, could you present the convergence of ELAN, compared with SwinIR and other CNN-based models? It can provide a more comprehensive comparison and better show the advantages of ELAN.
Thanks a lot.
BTW, the neat model architecture is definitely appealing.
As described in the title, how can I get the visual results on Set5? I didn`t find the corresponding code in the program.
div2k_cache是干嘛的
Hello,
Thanks for your great work, an efficient and neat transformer framework is essential for low-level vision I think.
According to your work, I tried discard the attention mask and positional encoding in SwinIR, the training and inference speed is largely improved, and the attention mask has slight effect on performance. However, the performance severely droped after removing RPE in original SwinIR.
Could you please give me some hints about how can we discard RPE (and attention mask) correctly? directly removing codes related to positional encoding or removing positional encoding should incorporate someother necessary elements?
Looking forward to your reply, thanks~
Hello,
MSU Graphics & Media Lab Video Group has recently launched two new Super-Resolution Benchmarks.
If you are interested in participating, you can add your algorithm following the submission steps:
We would be grateful for your feedback on our work!
Hi,
I am unable to download dataset as it requires BaiduNetdisk account and outside china it is not working. Can you please share the drive link?
We transferred the model to onnx and found that the accuracy was not consistent with the original model
Excellent work!
I hope can get the pretrain model and further research it.
Thanks~
CODA PACAGE HAVE TWO UTILS.PY
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.