search

Markfryazino / wav2lip-hq

Extension of Wav2Lip repository for processing high-quality videos.
534 stars 236 forks source link

Does anyone have any idea why the output video has a mask on? Im using the yunying model instead of the max one, maybe thats the issue? #35

Open ImBadAtNames2019 opened 1 year ago

ImBadAtNames2019 commented 1 year ago

Screenshot 2023-02-01 230909

GucciFlipFlops1917 commented 1 year ago

It appears to be the case. The weird mask disappears only after entering in '--no_sr'. You can try my fork if you like. Includes an ESRGAN model link particularized to faces and trained on a larger dataset than the particularized example in this repo.

ImBadAtNames2019 commented 1 year ago

It appears to be the case. The weird mask disappears only after entering in '--no_sr'. You can try my fork if you like. Includes an ESRGAN model link particularized to faces and trained on a larger dataset than the particularized example in this repo.

Thanks, i will check it out.