Predicting Head Pose in Dyadic Conversation

Greenwood, David, Laycock, Stephen and Matthews, Iain (2017) Predicting Head Pose in Dyadic Conversation. In: International Conference on Intelligent Virtual Agents. Intelligent Virtual Agents, 10498 . Springer, pp. 160-169. ISBN 978-3-319-67400-1

[img]
Preview
PDF (Accepted manuscript) - Submitted Version
Download (837kB) | Preview

    Abstract

    Natural movement plays a significant role in realistic speech animation. Numerous studies have demonstrated the contribution visual cues make to the degree we, as human observers, find an animation acceptable. Rigid head motion is one visual mode that universally co-occurs with speech, and so it is a reasonable strategy to seek features from the speech mode to predict the head pose. Several previous authors have shown that prediction is possible, but experiments are typically confined to rigidly produced dialogue. Expressive, emotive and prosodic speech exhibit motion patterns that are far more difficult to predict with considerable variation in expected head pose. People involved in dyadic conversation adapt speech and head motion in response to the others’ speech and head motion. Using Deep Bi-Directional Long Short Term Memory (BLSTM) neural networks, we demonstrate that it is possible to predict not just the head motion of the speaker, but also the head motion of the listener from the speech signal.

    Item Type: Book Section
    Faculty \ School: Faculty of Science > School of Computing Sciences
    Related URLs:
    Depositing User: Pure Connector
    Date Deposited: 13 Sep 2017 06:08
    Last Modified: 09 Apr 2019 10:47
    URI: https://ueaeprints.uea.ac.uk/id/eprint/64845
    DOI: 10.1007/978-3-319-67401-8_18

    Actions (login required)

    View Item