TR2017-008

Attention-Based Multimodal Fusion for Video Description


    •  Hori, C., Hori, T., Lee, T.-Y., Sumi, K., Hershey, J.R., Marks, T.K., "Attention-Based Multimodal Fusion for Video Description", arXiv, January 2017.
      BibTeX Download PDF
      • @techreport{MERL_TR2017-008,
      • author = {Hori, C. and Hori, T. and Lee, T.-Y. and Sumi, K. and Hershey, J.R. and Marks, T.K.},
      • title = {Attention-Based Multimodal Fusion for Video Description},
      • institution = {MERL - Mitsubishi Electric Research Laboratories},
      • address = {Cambridge, MA 02139},
      • number = {TR2017-008},
      • month = jan,
      • year = 2017,
      • url = {http://www.merl.com/publications/TR2017-008/}
      • }
  • MERL Contacts:
  • Research Areas:

    Computer Vision, Machine Learning, Multimedia, Speech & Audio

  • Research Highlights:
    Research Highlight
    Video Description

Currently successful methods for video description are based on encoder-decoder sentence generation using recurrent neural networks (RNNs). Recent work has shown the advantage of integrating temporal and/or spatial attention mechanisms into these models, in which the decoder network predicts each word in the description by selectively giving more weight to encoded features from specific time frames (temporal attention) or to features from specific spatial regions (spatial attention). In this paper, we propose to expand the attention model to selectively attend not just to specific times or spatial regions, but to specific modalities of input such as image features, motion features, and audio features. Our new modality-dependent attention mechanism, which we call multimodal attention, provides a natural way to fuse multimodal information for video description. We evaluate our method on the Youtube2Text dataset, achieving results that are competitive with current state of the art. More importantly, we demonstrate that our model incorporating multimodal attention as well as temporal attention significantly outperforms the model that uses temporal attention alone.