The PSNR metrics of different tasks

liu-vis / DualResidualNetworks

Dual Residual Networks Leveraging the Potential of Paired Operations for Image Restoration

MIT License

154 stars 24 forks source link

The PSNR metrics of different tasks #16

Closed LiuzhuForFun closed 4 years ago

LiuzhuForFun commented 4 years ago

Hi, Dr Liu Thanks for your great job! I check the testing codes provided by this work， different tasks use different PSNR computation. E.g. (1) Denosing: the average channels of PSNR (2) Deraining Drop: use the Y channel of images. Could you please tell me that the reason about these implementations? Thanks a lot!

liu-vis commented 4 years ago

Hello Mr. Liu, Thank you for your interest in our work! :) I chose the way of computing PSNR following the previous work(s) for a task, e.g.,

For denoising, I follow the paper "Exploiting the Potential of Standard Convolutional Autoencoders for Image Restoration by Evolutionary Search" (ICML18, the SOTA at that time)
For de-raindrop, I follow "Attentive Generative Adversarial Network for Raindrop Removal from A Single Image" (CVPR'18, the SOTA at the time)

LiuzhuForFun commented 4 years ago

Thanks for your attention!