Gaze Augmentation in Egocentric Video Improves Awareness of Intention
summary
Egocentric video with a superimposed gaze point improves people’s ability to predict a driver’s turn: observers view 15 driving clips recorded with a city car simulator and an ergonair decouplis binoculars 60 Hz gaze tracker, and 12 participants guess direction at intersections. Accuracy rises from a median 9/15 without gaze to 12/15 with gaze, and confidence increases significantly. Results point to real, measurable value for gaze augmentation in intention prediction and suggest follow-up studies in true collaborative settings.
AI-written summary · about
chapters (8)
- 0:00Introduction and motivation
- 2:53Previous research and motivation
- 5:23Hypothesis and gaze behavior
- 8:13Study design and driving scenario
- 10:23Participant task and setup
- 12:10Results and findings
- 14:41Implications and future work
- 15:46Conclusion and Q&A
related talks
09mThe Subconscious Director: Dynamically Personalizing Videos Using Gaze Data
09mGazeChat: Enhancing Virtual Conferences with Gaze Awareness and Interactive 3D Photos
14mTransform intent into action through your eyes
22mEye tracking the user experience – should you?
17mGaze-Assisted Typing for Smart Glasses
11mWatching 360° Videos Together
34mHow Eyetracking can be Beginning of the End of Privacy
18mGaze-Driven Video Delivery: Science Fiction or Viable Link to the Metaverse?