zhanglonghao1992 / One-Shot_Free-View_Neural_Talking_Head_Synthesis

Pytorch implementation of paper "One-Shot Free-View Neural Talking-Head Synthesis for Video Conferencing"
Other
735 stars 141 forks source link

大佬您好!运行demo的时候beta和new版本结果都可以跑出结果,但是这两个结果视频却没有声音,这个问题您有解决办法么~ #40

Closed czh007 closed 2 years ago

czh007 commented 2 years ago

运行beta版本的命令是: !python demo.py --result_video data/result.mp4 --config config/vox-256-beta.yaml --checkpoint checkpoint/15kp-ep119.pth.tar --source_image data/man.png --driving_video data/man.mp4 --relative --adapt_scale --find_best_frame --gen original --yaw 20 --pitch 0 --roll 0

运行spade版本的命令是: !python demo.py --result_video data/result.mp4 --config config/vox-256-spade.yaml --checkpoint checkpoint/00000189-checkpoint.pth.tar --source_image data/man.png --driving_video data/man.mp4 --relative --adapt_scale --find_best_frame --free_view --yaw 20 --pitch 0 --roll 0

图片和有声视频输入数据是: ironman

https://user-images.githubusercontent.com/38771857/149393660-9f71b6b3-84cd-4a0d-8ce6-9def73e5c89f.mp4

输出的两个视频数据(beta,new):

https://user-images.githubusercontent.com/38771857/149393837-62d6e84b-7e5d-4af0-9842-44b79aad03b1.mp4

https://user-images.githubusercontent.com/38771857/149393840-31e26daa-0405-42c1-beeb-f5106e36690b.mp4