ReasonCD: A Multimodal Reasoning Large Model for Implicit Change-of-Interest Semantic Mining

TL;DR

ReasonCD is a multimodal large model that effectively detects changes in remote sensing images by understanding implicit user intents through advanced reasoning, outperforming existing methods especially when explicit descriptions are unavailable.

Contribution

The paper introduces ReasonCD, a novel multimodal reasoning model that mines implicit change-of-interest semantics using large language models, enhancing remote sensing change detection accuracy.

Findings

Achieves 92.1% F1 score on BCDD dataset.

Excels at reasoning-based change detection tasks.

Can explain its reasoning process to assist human decisions.

Abstract

Remote sensing image change detection is one of the fundamental tasks in remote sensing intelligent interpretation. Its core objective is to identify changes within change regions of interest (CRoI). Current multimodal large models encode rich human semantic knowledge, which is utilized for guidance in tasks such as remote sensing change detection. However, existing methods that use semantic guidance for detecting users' CRoI overly rely on explicit textual descriptions of CRoI, leading to the problem of near-complete performance failure when presented with implicit CRoI textual descriptions. This paper proposes a multimodal reasoning change detection model named ReasonCD, capable of mining users' implicit task intent. The model leverages the powerful reasoning capabilities of pre-trained large language models to mine users' implicit task intents and subsequently obtains different change detection results based on these intents. Experiments on public datasets demonstrate that the model achieves excellent change detection performance, with an F1 score of 92.1\% on the BCDD dataset. Furthermore, to validate its superior reasoning functionality, this paper annotates a subset of reasoning data based on the SECOND dataset. Experimental results show that the model not only excels at basic reasoning-based change detection tasks but can also explain the reasoning process to aid human decision-making.