Dynamic Stereoscopic 3D Parameter Adjustment for Enhanced Depth Discrimination
summary
Two realtime algorithms adjust separation and convergence: one computes a low-resolution depth map, the other uses gaze from a Tobii EyeX to target the fixated object. Implementation runs in Unity on a GTX 780 driving a BenQ 3D monitor with wired NVIDIA glasses; study with 36 participants compares static, dynamic, and eye-tracked modes across three tasks (depth ranking, relative depth, path tracing). Eye-tracked dynamic yields the best accuracy and lowest error magnitude, dynamic alone helps when no near object, and only path tracing shows faster response times.
AI-written summary · about
chapters (18)
- 0:00Introduction
- 0:58Motivation
- 2:58Research Goal
- 3:25Related Work
- 5:11Stereoscopic Algorithms
- 6:07Dynamic Stereo Algorithm
- 6:32Stereo Parameter Table
- 7:04Eye-tracked Dynamic Stereo
- 8:30User Evaluation
- 9:07Experimental Tasks
- 10:10Study Hypotheses
- 10:39Subjects and Apparatus
- 11:10Depth Ranking Task
- 12:05Relative Depth Judgment Task
- 12:54Path Tracing Task
- 13:53Discussion
- 16:53Conclusion
- 17:37Acknowledgements
related talks
18mDepth Conflict Reduction for Stereo VR Video Interfaces
07mBringing back human interaction to a digital world
16mInteractive 360-Degree Glasses-Free Tabletop 3D Display
10mImpostor-based Rendering Acceleration for Virtual, Augmented, and Mixed Reality
43mVirtual Reality; Past, Present and Future
22mPinpointing: Precise Head- and Eye-Based Target Selection for Augmented Reality
17mMise-Unseen: Using Eye Tracking to Hide Virtual Reality Scene Changes in Plain Sight
30mLeveraging Decades of Research in 3D UI Makes or Breaks XR