search

cswry / SeeSR

[CVPR2024] SeeSR: Towards Semantics-Aware Real-World Image Super-Resolution
Apache License 2.0
328 stars 14 forks source link

FID problem #45

Open striveAgain opened 1 month ago

striveAgain commented 1 month ago

Hello, thank you for sharing the code of SeeSR! That's an amazing work indeed!

When I tried to calculate the metric FID using the code proposed in basicsr/metrics/fid.py, the result was quite confusing.

Here is how I calculate the metric FID. If any problem, feel free to point out. image

First, I defined a function which outputs a list composed of all images. image Then, I processed SR images and GT images, respectively, using inception_v3 to extract features and calculating FID.

As for the results, when the param normalize_input set to False, I got FID = 118.71415109991415. And when the param normalize_input set to True, I got FID = 126.1874280351401.

Both of them are quite higher than the paper mentioned, this makes me realize which step seems to have gone wrong. Could you please give me some advice? And whether the param normalize_input should be set to True or not?

Thanks a lot!

cswry commented 1 month ago

Hello, I use pyiqa for metrics calculating, please follow the instruction of IQA-PyTorch