Seg2Change: Adapting Open-Vocabulary Semantic Segmentation Model for Remote Sensing Change Detection

TL;DR

Seg2Change is a framework that adapts open-vocabulary semantic segmentation models for remote sensing change detection, enabling detection across arbitrary categories with state-of-the-art performance.

Contribution

The paper introduces a novel adapter, Seg2Change, and a category-agnostic change detection dataset, CA-CDD, for open-vocabulary change detection in remote sensing imagery.

Findings

Achieves +9.52 IoU on WHU-CD dataset.

Achieves +5.50 mIoU on SECOND dataset.

Outperforms existing methods in open-vocabulary change detection.

Abstract

Change detection is a fundamental task in remote sensing, aiming to quantify the impacts of human activities and ecological dynamics on land-cover changes. Existing change detection methods are limited to predefined classes in training datasets, which constrains their scalability in real-world scenarios. In recent years, numerous advanced open-vocabulary semantic segmentation models have emerged for remote sensing imagery. However, there is still a lack of an effective framework for directly applying these models to open-vocabulary change detection (OVCD), a novel task that integrates vision and language to detect changes across arbitrary categories. To address these challenges, we first construct a category-agnostic change detection dataset, termed CA-CDD. Further, we design a category-agnostic change head to detect the transitions of arbitrary categories and index them to specific classes. Based on them, we propose Seg2Change, an adapter designed to adapt open-vocabulary semantic segmentation models to change detection task. Without bells and whistles, this simple yet effective framework achieves state-of-the-art OVCD performance (+9.52 IoU on WHU-CD and +5.50 mIoU on SECOND). Our code is released at https://github.com/yogurts-sy/Seg2Change.