High-Fidelity Lake Extraction via Two-Stage Prompt Enhancement: Establishing a Novel Baseline and Benchmark

TL;DR

This paper presents LEPrompter, a two-stage prompt enhancement framework for high-fidelity lake extraction from remote sensing images, establishing a new baseline and benchmark with improved accuracy and efficiency.

Contribution

It introduces a novel prompt-based dataset construction and a two-stage prompt enhancement framework for more accurate lake segmentation from remote sensing imagery.

Findings

Significant performance improvements over previous methods.

Effective automated lake extraction without extra parameters or GFlops.

Robustness to varied lake shapes and data noise.

Abstract

Lake extraction from remote sensing imagery is a complex challenge due to the varied lake shapes and data noise. Current methods rely on multispectral image datasets, making it challenging to learn lake features accurately from pixel arrangements. This, in turn, affects model learning and the creation of accurate segmentation masks. This paper introduces a prompt-based dataset construction approach that provides approximate lake locations using point, box, and mask prompts. We also propose a two-stage prompt enhancement framework, LEPrompter, with prompt-based and prompt-free stages during training. The prompt-based stage employs a prompt encoder to extract prior information, integrating prompt tokens and image embedding through self- and cross-attention in the prompt decoder. Prompts are deactivated to ensure independence during inference, enabling automated lake extraction without introducing additional parameters and GFlops. Extensive experiments showcase performance improvements of our proposed approach compared to the previous state-of-the-art method. The source code is available at https://github.com/BastianChen/LEPrompter.