Related Experiment Video
Updated: Jun 10, 2025
06:53
Creating Virtual-hand and Virtual-face Illusions to Investigate Self-representation
Published on: March 1, 2017
13.2K
Assessment of autostereoscopic perception using artificial intelligence-enhanced face tracking technology
1Department of Optometry, The Second Hospital of Jilin University, Changchun, China.
Plos One
|October 17, 2024
Summary
Artificial intelligence (AI)-enhanced autostereoscopic technology shows promise for accurate stereopsis measurements. This AI-powered tablet ensures stable head positioning, improving depth perception assessments in clinical settings.
Area of Science:
- Ophthalmology
- Computer Science
- Human-Computer Interaction
Background:
- Stereopsis, the perception of depth from binocular vision, is crucial for autostereoscopic displays.
- Maintaining stable head position is a significant challenge in traditional stereopsis assessments.
- Artificial intelligence (AI)-enhanced face tracking offers a potential solution for consistent visual stimuli delivery.
Purpose of the Study:
- To evaluate the effectiveness of AI-enhanced face tracking in autostereoscopic technology for stereopsis measurement.
- To assess the inter-session reliability and agreement of an AI-powered autostereoscopic tablet with conventional stereotests.
Main Methods:
- An autostereoscopic tablet (Lume Pad 2) with AI-enhanced face tracking simulated Stereo Fly and TNO Stereotests.
- Inter-session reliability was assessed in 30 children and 30 adults.
- Agreement with conventional methods was evaluated in 181 children and 160 adults using weighted Kappa and Bland-Altman analysis.
Main Results:
- The AI-powered tablet demonstrated high inter-session reliability (κ > 0.80) across most tests.
- Moderate reliability was observed for the TNO Stereotest in adults (κ = 0.571).
- Significant agreement was found between AI-based and conventional methods, with minimal differences.
Conclusions:
- AI-integrated autostereoscopic technology provides reliable and accurate clinical stereopsis measurements.
- Sub-pixel precision in AI-enhanced displays enhances the potential for precise depth perception evaluations.
