Related Experiment Video
Updated: Apr 10, 2026
06:54
Photorealistic Learned Landscapes for Augmented Reality
Published on: June 27, 2025
888
An Augmented Reality Study for Enhancing Spatial-Temporal Cognition in Indoor Emergency Response.
IEEE Transactions on Visualization and Computer Graphics
|April 8, 2026
Summary
Augmented Reality (AR) maps show task-dependent performance for indoor emergency navigation. Egocentric AR views may improve time estimation, while 2D/3D maps excel at distance judgments for first responders.
Area of Science:
- Human-Computer Interaction
- Cognitive Science
- Emergency Response Technologies
Background:
- Effective spatial-temporal cognition is critical for indoor emergency situational awareness.
- Complex indoor environments with occlusions challenge first responders' judgment of distances, scales, and task times.
- Augmented Reality (AR) offers potential solutions for enhancing these cognitive abilities.
Purpose of the Study:
- To investigate the impact of different AR visualization modes on spatial and temporal estimation tasks for indoor emergency scenarios.
- To compare the performance of 2D Map, 3D Map, and Egocentric Map visualizations within a mixed reality environment.
- To determine the optimal AR visualization strategy for specific spatial and temporal cognitive tasks.
Main Methods:
- A digital twin of a three-level indoor structure was created using LiDAR scans and reality capturing technologies.
- A mixed reality environment was developed using the Unity engine and Microsoft HoloLens2.
- A user study (N=30) evaluated performance across 2D Map, 3D Map, and Egocentric Map conditions for spatial and temporal estimation tasks.
Main Results:
- Egocentric maps showed a trend towards lower error in Estimated Time of Arrival (ETA) tasks (MAPE = 17.2%) compared to 2D maps (MAPE = 23.3%), though not statistically significant after corrections.
- 2D and 3D maps demonstrated significantly lower error in distance-between-objects tasks (MAPE = 24.3% and 17.7%) compared to egocentric views (MAPE = 36.8%, p_adj < 0.01).
- A task-dependent trade-off was observed, indicating no single visualization is universally superior.
Conclusions:
- AR visualization modes exhibit task-specific effectiveness in spatial-temporal cognition for emergency response.
- Egocentric AR views may enhance temporal cognition for tasks like ETA prediction.
- Exocentric (2D/3D) views remain superior for precise spatial judgments, such as distance estimation.
