v0.0.2-rc
版本发布时间: 2023-06-05 22:44:03
OpenTalker/SadTalker最新发布版本:v0.0.2-rc(2023-06-05 22:44:03)
Multiple new optimizations for SadTalker:
- A new
option
ofsize
can be used to choose the model, (256 or 512) currently. eg,python inference.py --size 512
- A new 512x512 face render model is released, which may show better results in the month open and higher resolution.
- set new preprocess:
extcrop
andextfull
, where the face crop is larger. eg,python inference.py --preprocess extcrop
- We package nearly all the models using
safetensor
and remove some necessary weights: checkpoint packages from7GB -> 1GB
! - Reduce the requirement of installation, now,
dlib
orface-alignment
is not a necessary package for a generation! - Since we have not trained the new mappingnet, we still use two different mappingnet as the previous release. It will be merged into a unified safetensors in the future.
- add a script of
webui.sh
for Linux. - Multiple
webui
functions are added:pose style
,face model resolution
,batch size
, etc. - Merge PR of #303 to boost the speed of face enhancers. Thanks for this contribution! very cool!
1、 SadTalker_V0.0.2_256.safetensors 691.48MB
2、 SadTalker_V0.0.2_512.safetensors 691.48MB
3、 mapping_00229-model.pth.tar 148.32MB
4、 mapping_00109-model.pth.tar 148.56MB