You said that?

Joon Son Son, Amir Jamaludin and Andrew Zisserman

Abstract

We present a method for generating a video of a talking face. The method takes as inputs: (i) still images of the target face, and (ii) an audio speech segment; and outputs a video of the target face lip synched with the audio. The method runs in real time and is applicable to faces and audio not seen at training time. To achieve this we propose an encoder-decoder CNN model that uses a joint embedding of the face and audio to generate synthesised talking face video frames. The model is trained on tens of hours of unlabelled videos. We also show results of re-dubbing videos using speech from a different person.

Session

Orals - Face Analysis

Files

Paper (PDF)
Supplementary (PDF)

DOI

10.5244/C.31.109
https://dx.doi.org/10.5244/C.31.109

Citation

Bibtex

            @inproceedings{BMVC2017_109,
                title={You said that?},
                author={Joon Son Son, Amir Jamaludin and Andrew Zisserman},
                year={2017},
                month={September},
                pages={109.1-109.12},
                articleno={109},
                numpages={12},
                booktitle={Proceedings of the British Machine Vision Conference (BMVC)},
                publisher={BMVA Press},
                editor={Tae-Kyun Kim, Stefanos Zafeiriou, Gabriel Brostow and Krystian Mikolajczyk},
                doi={10.5244/C.31.109},
                isbn={1-901725-60-X},
                url={https://dx.doi.org/10.5244/C.31.109}
            }