ChangeChat: An Interactive Model for Remote Sensing Change Analysis via Multimodal Instruction Tuning
This provides an interactive solution for remote sensing professionals to analyze Earth's dynamic processes, addressing limitations in traditional change detection and recent change captioning methods.
The paper tackles the problem of remote sensing change analysis by introducing ChangeChat, a bitemporal vision-language model that supports interactive queries like change captioning, quantification, and localization, achieving performance comparable to or better than SOTA methods on specific tasks and significantly surpassing GPT-4.
Remote sensing (RS) change analysis is vital for monitoring Earth's dynamic processes by detecting alterations in images over time. Traditional change detection excels at identifying pixel-level changes but lacks the ability to contextualize these alterations. While recent advancements in change captioning offer natural language descriptions of changes, they do not support interactive, user-specific queries. To address these limitations, we introduce ChangeChat, the first bitemporal vision-language model (VLM) designed specifically for RS change analysis. ChangeChat utilizes multimodal instruction tuning, allowing it to handle complex queries such as change captioning, category-specific quantification, and change localization. To enhance the model's performance, we developed the ChangeChat-87k dataset, which was generated using a combination of rule-based methods and GPT-assisted techniques. Experiments show that ChangeChat offers a comprehensive, interactive solution for RS change analysis, achieving performance comparable to or even better than state-of-the-art (SOTA) methods on specific tasks, and significantly surpassing the latest general-domain model, GPT-4. Code and pre-trained weights are available at https://github.com/hanlinwu/ChangeChat.