Urban Waterlogging Detection: A Challenging Benchmark and Large-Small Model Co-Adapter

TL;DR

This paper introduces a new challenging benchmark for urban waterlogging detection and proposes a novel Large-Small Model co-adapter framework that leverages large and small models for improved detection under adverse conditions.

Contribution

It establishes a comprehensive urban waterlogging benchmark and proposes a co-adapter paradigm combining large and small models with novel modules for enhanced detection performance.

Findings

The benchmark presents diverse adverse conditions for waterlogging detection.

The LSM-adapter outperforms existing methods on the benchmark.

The proposed modules improve adaptation to challenging environmental conditions.

Abstract

Urban waterlogging poses a major risk to public safety and infrastructure. Conventional methods using water-level sensors need high-maintenance to hardly achieve full coverage. Recent advances employ surveillance camera imagery and deep learning for detection, yet these struggle amidst scarce data and adverse environmental conditions. In this paper, we establish a challenging Urban Waterlogging Benchmark (UW-Bench) under diverse adverse conditions to advance real-world applications. We propose a Large-Small Model co-adapter paradigm (LSM-adapter), which harnesses the substantial generic segmentation potential of large model and the specific task-directed guidance of small model. Specifically, a Triple-S Prompt Adapter module alongside a Dynamic Prompt Combiner are proposed to generate then merge multiple prompts for mask decoder adaptation. Meanwhile, a Histogram Equalization Adap-ter module is designed to infuse the image specific information for image encoder adaptation. Results and analysis show the challenge and superiority of our developed benchmark and algorithm. Project page: \url{https://github.com/zhang-chenxu/LSM-Adapter}