search

Snowfallingplum / SSAT

[AAAI 2022] This is the official pytorch code for SSAT: A Symmetric Semantic-Aware Transformer Network for Makeup Transfer and Removal
84 stars 20 forks source link

Generate artifacts that appear in the final image #8

Open YifanWangCN opened 2 years ago

YifanWangCN commented 2 years ago

Thank you very much for your work and the improved generalization robustness of this algorithm, but I would still like to ask you what causes the phenomenon shown in the figure. artifacts

Snowfallingplum commented 2 years ago

I presume this may be caused by image cropping, you can crop the input image so that the face is larger like the reference image. This phenomenon is caused by the large gap between the distribution of training data and input data.

YifanWangCN commented 2 years ago

I presume this may be caused by image cropping, you can crop the input image so that the face is larger like the reference image. This phenomenon is caused by the large gap between the distribution of training data and input data.

It truly work, thanks.