search

EvelynFan / FaceFormer

[CVPR 2022] FaceFormer: Speech-Driven 3D Facial Animation with Transformers
MIT License
778 stars 133 forks source link

There is no sound of the rendered videos in the demo/output ?why? #53

Open wzx7084 opened 1 year ago

FangmingTian commented 1 year ago

Sorry I can't help with your problem now since I haven't run the code. By the way, could you help with #52 issued sent by Tianfangming? The issue is that I can't find these four files: data_verts.npy, raw_audio_fixed.pkl, templates.pkl and subj_seq_to_idx.pkl. It'll be a big favor if you could offer some help!

wzx7084 commented 1 year ago

You can download them on https://voca.is.tue.mpg.de/. Click the Download button at the top to go to the new screen,you'll see training data at the bottom of the page. data_verts.npy, raw_audio_fixed.pkl, templates.pkl and subj_seq_to_idx.pkl in this zip.

FangmingTian commented 1 year ago

Couldn't be more grateful! #

You can download them on https://voca.is.tue.mpg.de/. Click the Download button at the top to go to the new screen,you'll see training data at the bottom of the page. data_verts.npy, raw_audio_fixed.pkl, templates.pkl and subj_seq_to_idx.pkl in this zip.

Shan924 commented 1 year ago

@wzx7084 In demo.py, find cmd = ('ffmpeg' + ' -i {0} -pix_fmt yuv420p -qscale {1} '.format( tmp_video_file.name, video_fname)).split(), replace it with cmd = ('ffmpeg' + ' -i {0} -i {1} -pix_fmt yuv420p -qscale 0 -vcodec copy -acodec aac {2} '.format(tmp_video_file.name,wav_path, video_fname)).split()

xmaple04 commented 1 year ago

@wzx7084 In demo.py, find cmd = ('ffmpeg' + ' -i {0} -pix_fmt yuv420p -qscale {1} '.format( tmp_video_file.name, video_fname)).split(), replace it with cmd = ('ffmpeg' + ' -i {0} -i {1} -pix_fmt yuv420p -qscale 0 -vcodec copy -acodec aac {2} '.format(tmp_video_file.name,wav_path, video_fname)).split()

Thank you very much. It works fine for me.

NicholasLiuSc commented 1 year ago

What's wrong with rendered videos without faces?