output resolution problem

[labels168]

This is a great open source, but there is a gap between the resolution of the output and the original video, the resolution of the output is low, and there are jagged, is there a way to deal with it, thank you very much!

[indianajson]

I would also be interested in a way to improve the output resolution when the system has the necessary RAM + GPU power to do so. There does not appear to be any information about this in the documentation.

[kunncheng]

You can try to use GFPGAN/GPEN again or other super-resolution/face restoration methods to enhance the generated video.
It's worth noting that GFPGAN will have a slightly change on identity.

[fjesikfjdskl]

When I used GPEN to enhance the face, problems such as facial mosaics appeared in the result of synthesis. Have you ever had such problems

[AIhasArrived]

I would also be interested in a way to improve the output resolution when the system has the necessary RAM + GPU power to do so. There does not appear to be any information about this in the documentation.

Hello @indianajson did you notice this tech used all the RAM and very few of GPU??

Do you know how to solve this?

[mahnoorfawad]

You can try to use GFPGAN/GPEN again or other super-resolution/face restoration methods to enhance the generated video. It's worth noting that GFPGAN will have a slightly change on identity.

how to use the super-resolution sr_model on colab, it doesnt download the RealESR model or make a srmodel directory