search

Zhangjinso / PISE

123 stars 28 forks source link

how to visualize the parsing map? #38

Closed Jerric-wu closed 5 months ago

Zhangjinso commented 2 years ago

You can refer to that

Jerric-wu commented 2 years ago

You can refer to that Thank you very much for your reply, I still have a question is that how to calculate the LPIPS. What do the ground-truth images and the generated images referring? Are the generated images from training or the 'eval_results', if come from 'eval_result', what are the corresponding ground-truth images? Look forward to your reply, thanks.