Self-Supervised Learning for Identifying Defects in Sewer Footage

TL;DR

This paper presents a self-supervised learning approach for sewer defect detection that reduces the need for labeled data and model size, offering a scalable and cost-effective solution for infrastructure maintenance.

Contribution

It introduces a novel SSL application for sewer inspection, achieving competitive results with smaller models and less data than traditional supervised methods.

Findings

Model is at least 5 times smaller than existing approaches.

Achieves competitive performance with only 10% of labeled data.

Highlights SSL's potential for resource-limited sewer maintenance.

Abstract

Sewerage infrastructure is among the most expensive modern investments requiring time-intensive manual inspections by qualified personnel. Our study addresses the need for automated solutions without relying on large amounts of labeled data. We propose a novel application of Self-Supervised Learning (SSL) for sewer inspection that offers a scalable and cost-effective solution for defect detection. We achieve competitive results with a model that is at least 5 times smaller than other approaches found in the literature and obtain competitive performance with 10\% of the available data when training with a larger architecture. Our findings highlight the potential of SSL to revolutionize sewer maintenance in resource-limited settings.