DeltaVLM：通过指令引导的差异感知实现交互式遥感影像变化分析
DeltaVLM: Interactive Remote Sensing Image Change Analysis via Instruction-guided Difference Perception
AI总结：
本研究提出遥感影像变化分析新范式，构建ChangeChat-105k数据集，推出DeltaVLM架构，通过差异感知与指令对齐实现交互式变化分析，性能达当前最优。
摘要：
多时相卫星影像中土地覆盖变化的准确解译对现实场景至关重要。然而，现有方法通常仅提供一次性变化掩码或静态描述，限制了其支持交互式、查询驱动分析的能力。本研究提出遥感影像变化分析（RSICA）这一新范式，结合变化检测与视觉问答的优势，实现对双时相遥感影像中变化的多轮、指令引导式探索。为支撑该任务，我们构建了大规模指令遵循数据集ChangeChat-105k，该数据集通过规则与GPT辅助的混合流程生成，覆盖六种交互类型：变化描述、分类、量化、定位、开放式问答及多轮对话。基于该数据集，我们提出专为交互式RSICA设计的端到端架构DeltaVLM。DeltaVLM具备三项创新：（1）经微调的双时相视觉编码器，用于捕捉时间差异；（2）带有跨语义关系度量（CSRM）机制的视觉差异感知模块，用于解译变化；（3）指令引导的Q-former，可有效从视觉变化中提取与查询相关的差异信息，并将其与文本指令对齐。我们使用冻结的大语言模型在ChangeChat-105k上训练DeltaVLM，仅调整视觉与对齐模块以优化效率。大量实验与消融研究表明，DeltaVLM在单轮描述与多轮交互式变化分析任务上均达到最优性能，优于现有的多模态大语言模型与遥感视觉语言模型。代码、数据集与预训练权重可在https://github.com/hanlinwu/DeltaVLM获取。
英文摘要：
Accurate interpretation of land-cover changes in multi-temporal satellite imagery is critical for real-world scenarios. However, existing methods typically provide only one-shot change masks or static captions, limiting their ability to support interactive, query-driven analysis. In this work, we introduce remote sensing image change analysis (RSICA) as a new paradigm that combines the strengths of change detection and visual question answering to enable multi-turn, instruction-guided exploration of changes in bi-temporal remote sensing images. To support this task, we construct ChangeChat-105k, a large-scale instruction-following dataset, generated through a hybrid rule-based and GPT-assisted process, covering six interaction types: change captioning, classification, quantification, localization, open-ended question answering, and multi-turn dialogues. Building on this dataset, we propose DeltaVLM, an end-to-end architecture tailored for interactive RSICA. DeltaVLM features three innovations: (1) a fine-tuned bi-temporal vision encoder to capture temporal differences; (2) a visual difference perception module with a cross-semantic relation measuring (CSRM) mechanism to interpret changes; and (3) an instruction-guided Q-former to effectively extract query-relevant difference information from visual changes, aligning them with textual instructions. We train DeltaVLM on ChangeChat-105k using a frozen large language model, adapting only the vision and alignment modules to optimize efficiency. Extensive experiments and ablation studies demonstrate that DeltaVLM achieves state-of-the-art performance on both single-turn captioning and multi-turn interactive change analysis, outperforming existing multimodal large language models and remote sensing vision-language models. Code, dataset and pre-trained weights are available at https://github.com/hanlinwu/DeltaVLM.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
