question about RESIDE-Outdoor

google-research / maxim

[CVPR 2022 Oral] Official repository for "MAXIM: Multi-Axis MLP for Image Processing". SOTA for denoising, deblurring, deraining, dehazing, and enhancement.

https://arxiv.org/abs/2201.02973

Apache License 2.0

1.02k stars 110 forks source link

question about RESIDE-Outdoor #19

Closed YuZheng9 closed 2 years ago

YuZheng9 commented 2 years ago

Thank you for your work！

I have a little question. There are two versions of the RESIDE-outdoor dataset. RESIDE:V0 and RESIDE-β.

RESIDE:V0 has 313,950 synthetic hazy images. RESIDE-β has 72135 synthetic hazy images.

Which version of RESIDE-outdoor dataset did you use to train and get the results in the paper?

Thank you and look forward to your reply.

vztu commented 2 years ago

We used RESIDE:V0 (313k), as shown in Table 8 of https://arxiv.org/pdf/2201.02973.pdf