Researchers propose LTM, a multi-modal reconstruction framework that leverages outdated Digital Elevation Models (DEMs) as geometric priors for image-based 3D reconstruction in wildfire-prone regions. The method uses physics-based pixel-pixel alignment between images and DEM data to eliminate expensive feature matching procedures.

  • Utilizes legacy DEMs as geometric priors for large-scale terrain mapping.
  • Employs physics-based pixel-pixel alignment to reduce computational complexity.
  • Generates high-fidelity depth maps from posed images while maintaining real-time performance.
  • Validated using a large-terrain simulator based on a real wildfire-prone area.

The approach offers significant improvements in reconstruction accuracy and computational efficiency, providing a scalable solution for emergency response.