Instance Segmentation of Reinforced Concrete Bridges with Synthetic Point Clouds

TL;DR

This paper introduces a new synthetic data generation framework and a transformer-based model for accurate instance segmentation of bridge elements in point clouds, aiming to automate detailed bridge inspections.

Contribution

The study presents a novel synthetic data generation approach and applies a Mask3D transformer model with hyperparameter tuning and occlusion techniques for improved bridge element segmentation.

Findings

State-of-the-art performance on real LiDAR bridge point clouds

Effective synthetic data generation methods for training segmentation models

Demonstrated potential for automating detailed bridge inspections

Abstract

The National Bridge Inspection Standards require detailed element-level bridge inspections. Traditionally, inspectors manually assign condition ratings by rating structural components based on damage, but this process is labor-intensive and time-consuming. Automating the element-level bridge inspection process can facilitate more comprehensive condition documentation to improve overall bridge management. While semantic segmentation of bridge point clouds has been studied, research on instance segmentation of bridge elements is limited, partly due to the lack of annotated datasets, and the difficulty in generalizing trained models. To address this, we propose a novel approach for generating synthetic data using three distinct methods. Our framework leverages the Mask3D transformer model, optimized with hyperparameter tuning and a novel occlusion technique. The model achieves state-of-the-art performance on real LiDAR and photogrammetry bridge point clouds, respectively, demonstrating the potential of the framework for automating element-level bridge inspections.