2019 IEEE International Conference on Multimedia & Expo Workshops (ICMEW) /

We propose an end to end deep learning approach for generating real-time facial animation from just audio. Specifically, our deep architecture employs deep bidirectional long short-term memory network and attention mechanism to discover the latent representations of time-varying contextual informati...

Ausführliche Beschreibung

Gespeichert in:
Weitere Titel:
2019 IEEE International Conference on Multimedia & Expo Workshops
Körperschaft:
Institute of Electrical and Electronics Engineers (IEEE)
Format:
Elektronisch E-Book
Sprache:
Englisch
Veröffentlicht:
Piscataway, New Jersey : Institute of Electrical and Electronics Engineers (IEEE), 2019.
Zusammenfassung:
We propose an end to end deep learning approach for generating real-time facial animation from just audio. Specifically, our deep architecture employs deep bidirectional long short-term memory network and attention mechanism to discover the latent representations of time-varying contextual information within the speech and recognize the significance of different information contributed to certain face status. Therefore, our model is able to drive different levels of facial movements at inference and automatically keep up with the corresponding pitch and latent speaking style in the input audio, with no assumption or further human intervention. Evaluation results show that our method could not only generate accurate lip movements from audio, but also successfully regress the speaker's time-varying facial movements.
Umfang:
1 online resource : illustrations
ISBN:
9781538692141
1538692147
Schlagworte:
Multimedia systems > Congresses.
Three-dimensional display systems > Congresses.
Bezugswerke:
Links: