• Computer Science > Computer Vision and Pattern Recognition [Submitted on 23 Jan 2026] Title:Three-dimensional Damage Visualization of Civil Structures via Gaussian Splatting-enabled Digital Twins View PDFAbstract:Recent advancements in civil infrastructure inspections underscore the need for precise three-dimensional (3D) damage visualization on digital twins, transcending traditional 2D image-based damage identifications. • Compared to conventional photogrammetric 3D reconstruction techniques, modern approaches such as Neural Radiance Field (NeRF) and Gaussian Splatting (GS) excel in scene representation, rendering quality, and handling featureless regions. • Among them, GS stands out for its efficiency, leveraging discrete anisotropic 3D Gaussians to represent radiance fields, unlike NeRF’s continuous implicit model. • This study introduces a GS-enabled digital twin method tailored for effective 3D damage visualization. • The method’s key contributions include: 1) utilizing GS-based 3D reconstruction to visualize 2D damage segmentation results while reducing segmentation errors; 2) developing a multi-scale reconstruction strategy to balance efficiency and damage detail; 3) enabling digital twin updates as damage evolves over time. • Demonstrated on an open-source synthetic dataset for post-earthquake inspections, the proposed approach offers a promising solution for comprehensive 3D damage visualization in civil infrastructure digital twins.
Article Summaries:
- Computer Science > Computer Vision and Pattern Recognition [Submitted on 23 Jan 2026] Title:Three-dimensional Damage Visualization of Civil Structures via Gaussian Splatting-enabled Digital Twins View PDFAbstract:Recent advancements in civil infrastructure inspections underscore the need for precise three-dimensional (3D) damage visualization on digital twins, transcending traditional 2D image-based damage identifications. Compared to conventional photogrammetric 3D reconstruction techniques, modern approaches such as Neural Radiance Field (NeRF) and Gaussian Splatting (GS) excel in scene repr
Sources: