The Trauma THOMPSON Challenge Banner
Join

12 October 2023 - Thursday AM, Vancouver/CANADA

Registration: https://conferences.miccai.org/2023/en/

Keynote: Extended Reality for Medical Applications through Visual AI

Invited speaker: Helge Rhodin

Abstract:
Extended reality solutions capture and enhance the user's experience from an egocentric viewpoint. However, detecting, reconstructing, and making decisions based on body-worn or hand-held cameras poses significant challenges. In my presentation, I will briefly introduce my pioneering work in egocentric motion capture, as well as past and ongoing projects related to action recognition and body shape reconstruction. I will then delve into how I believe that these methods will reshape medical applications and how our latest advancements in neural character models enable telemedicine and AI-driven decision-making.

Biography:

Helge Rhodin is currently an Assistant Professor at UBC, leading a team of 10 graduate students and postdocs and is associated with the UBC computer vision and graphics labs. In 2024, he will start as full Professor at Bielefeld University, Germany, leading the Visual AI for Extended Reality group. Before, he was a lecturer and postdoc at EPFL and received his PhD from Saarland University, with his thesis on Motion Capture for Interactive Virtual Worlds. Rhodin's research interests range from fundamental 3D computer vision and self-supervised machine learning to applications in sports, medicine, neuroscience, and augmented reality.

Date: Thursday, October 12, 2023 (Times given in Pacific time zone)

Location:  Meeting Room 14, Vancouver Convention Center East Building Level 1

  • Introductory Talk by Organizers
    9:00 - 9:15: Introduction to The Trauma THOMPSON challenge. By Prof. Juan Wachs
  • Keynote: Extended Reality for Medical Applications through Visual AI
    9:15 - 10:00: By Prof. Helge Rhodin
  • Break
    10:00 - 10:30
  • Oral Session
    10:30 - 10:45: An ADG: Action Dictionary-guided learning model for action recognition and anticipation. By Trinh T.L. Voung  
    10:45 - 11:00: Video swim Transformer for action recognition and anticipation. By Guang Liang
  • 11:00 - 11:15: Task Oriented Network.  By Yupeng Zhuo  
  • 11:15 - 11:30: A frame-question cross attention for visual question answering task.  *By Trinh T.L. Voung  *
  • Winners and Awards
    11:30 - 11:50 By Kyle Couperus, MD, and Prof. Juan Wachs
  • Closing Remarks
    11:50 - 12:00 By Kyle Couperus, MD