TR2022-151

Human Perspective Scene Understanding via Multimodal Sensing

    •  Hori, C., "Human Perspective Scene Understanding via Multimodal Sensing," Tech. Rep. TR2022-151, Audio-Visual Scene Understanding Tutorial at CVPR 2021, June 2021.
      BibTeX TR2022-151 PDF Video
      • @techreport{Hori2021jun,
      • author = {Hori, Chiori},
      • title = {Human Perspective Scene Understanding via Multimodal Sensing},
      • institution = {Mitsubishi Electric Research Laboratories},
      • year = 2021,
      • month = jun,
      • url = {https://www.merl.com/publications/TR2022-151}
      • }
  • MERL Contact:
  • Research Areas:

    Artificial Intelligence, Computer Vision, Human-Computer Interaction, Speech & Audio

Abstract:

This talk introduces our research activities on scene-aware interaction including neural conversation, video captioning, Audio Visual Scene-aware Dialog (AVSD), and our demo for future car navigation using scene-aware interaction technologies.

Presented as part of a tutorial on Audio-Visual Scene Understanding, CVPR 2021.

 

  • Related Video