Text2Seg: Remote Sensing Image Semantic Segmentation via Text-Guided Visual Foundation Models

TL;DR

Text2Seg introduces a novel approach for remote sensing semantic segmentation that leverages visual foundation models and automatic prompt generation to reduce annotation dependency and improve zero-shot performance across diverse datasets.

Contribution

The paper presents Text2Seg, a new method that overcomes annotation limitations and enhances transferability in remote sensing segmentation by using visual foundation models and automatic prompts.

Findings

Significant improvement in zero-shot segmentation performance, with relative gains from 31% to 225%.

Reduces reliance on fully annotated datasets through automatic prompt generation.

Enhances generalization ability across diverse remote sensing datasets.

Abstract

Remote sensing imagery has attracted significant attention in recent years due to its instrumental role in global environmental monitoring, land usage monitoring, and more. As image databases grow each year, performing automatic segmentation with deep learning models has gradually become the standard approach for processing the data. Despite the improved performance of current models, certain limitations remain unresolved. Firstly, training deep learning models for segmentation requires per-pixel annotations. Given the large size of datasets, only a small portion is fully annotated and ready for training. Additionally, the high intra-dataset variance in remote sensing data limits the transfer learning ability of such models. Although recently proposed generic segmentation models like SAM have shown promising results in zero-shot instance-level segmentation, adapting them to semantic segmentation is a non-trivial task. To tackle these challenges, we propose a novel method named Text2Seg for remote sensing semantic segmentation. Text2Seg overcomes the dependency on extensive annotations by employing an automatic prompt generation process using different visual foundation models (VFMs), which are trained to understand semantic information in various ways. This approach not only reduces the need for fully annotated datasets but also enhances the model's ability to generalize across diverse datasets. Evaluations on four widely adopted remote sensing datasets demonstrate that Text2Seg significantly improves zero-shot prediction performance compared to the vanilla SAM model, with relative improvements ranging from 31% to 225%. Our code is available at https://github.com/Douglas2Code/Text2Seg.