A multimodal perception driven digital disaster site interactive display method
By integrating multimodal perception data and digital twin models, a fully closed-loop adaptive interactive display of digital disaster sites was achieved, solving the problems of single perception dimension and insufficient narrative, and enhancing the immersiveness and emotional resonance of the display.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- MIANYANG TEACHERS COLLEGE
- Filing Date
- 2026-04-28
- Publication Date
- 2026-07-21
AI Technical Summary
Existing digital heritage displays suffer from limited perceptual dimensions, lack of interactive loops, and insufficient attention to the unique characteristics of disaster narratives, failing to achieve multi-layered immersion, personalization, and emotional resonance in digital disaster heritage displays.
By deploying multiple sensors to synchronously collect data on the site environment, user behavior, and emotional perception, a multimodal fusion model is constructed for spatiotemporal alignment and feature fusion to generate a comprehensive perception state vector. Based on a high-precision digital twin model, multi-layer narrative rendering and adaptive interactive display are performed to achieve perception-driven adaptive interaction.
It achieves a fully closed-loop multimodal perception-driven system, breaking through the limitations of a single perception dimension, supporting multi-layered narrative rendering and emotional feedback, enhancing the immersiveness and emotional resonance of digital disaster site displays, and possessing self-evolution capabilities.
Smart Images

Figure CN122431528A_ABST