CIPHER: Culvert Inspection through Pairwise Frame Selection and High-Efficiency Reconstruction

TL;DR

This paper introduces a fast RGB-based 3D reconstruction pipeline for culvert inspection, combining frame pair selection and real-time geometry and semantics estimation to improve safety and efficiency.

Contribution

It presents a novel, efficient pipeline that selects informative frame pairs and performs real-time 3D reconstruction of culvert-like structures in repetitive environments.

Findings

Accurately reconstructs culvert geometries and depth maps

Enhances inspection efficiency with minimal human input

Operates in real-time with high accuracy

Abstract

Automated culvert inspection systems can help increase the safety and efficiency of flood management operations. As a key step to this system, we present an efficient RGB-based 3D reconstruction pipeline for culvert-like structures in visually repetitive environments. Our approach first selects informative frame pairs to maximize viewpoint diversity while ensuring valid correspondence matching using a plug-and-play module, followed by a reconstruction model that simultaneously estimates RGB appearance, geometry, and semantics in real-time. Experiments demonstrate that our method effectively generates accurate 3D reconstructions and depth maps, enhancing culvert inspection efficiency with minimal human intervention.