search

FuxiVirtualHuman / AAAI22-one-shot-talking-face

Code for One-shot Talking Face Generation from Single-speaker Audio-Visual Correlation Learning (AAAI 2022)
353 stars 61 forks source link

A bug in interface.py #2

Closed YifengMa9 closed 2 years ago

YifengMa9 commented 2 years ago

Line 74. there are extra space in keys. pose = [pos_data[" pose_Rx"][i], pos_data[" pose_Ry"][i], pos_data[" pose_Rz"][i],pos_data[" pose_Tx"][i], pos_data[" pose_Ty"][i], pos_data[" pose_Tz"][i]] should be pose = [pos_data["pose_Rx"][i], pos_data["pose_Ry"][i], pos_data["pose_Rz"][i],pos_data["pose_Tx"][i], pos_data["pose_Ty"][i], pos_data["pose_Tz"][i]]

Thanks for your remarkable work. The performance is amazing.

FuxiVirtualHuman commented 2 years ago

R

This is caused by the different versions of Openface and the different paltforms.

yaleimeng commented 2 years ago

windows版本的csv，列名前面有空格；Linux版本的csv文件，列名不包含空格。 项目代码是默认按前面带空格的形式，修改一下就可以了。

YifengMa9 commented 2 years ago

Thanks

rlgnswk commented 1 year ago

Thanks for your recording!