arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2511.02607cs.CVcs.CL
收藏

UniChange: 通过多模态大语言模型统一变化检测

UniChange: Unifying Change Detection with Multimodal Large Language Model

TMCC, Computer Science, Nankai University(天津大学计算机学院,南开大学);VCIP, Computer Science, Nankai University(南开大学计算机科学系);NKIARI, Futian, Shenzhen(深圳福田国家工程研究中心);CMEE, Sichuan Agricultural University(四川农业大学)

Xu Zhang, Danyang Li, Xiaohang Dong, Tianhao Wu, Hualong Yu, Jianye Wang, Qicheng Li, Xiang Li

AI总结:

UniChange通过多模态大语言模型统一变化检测任务,引入特殊标记和文本提示,实现BCD和SCD的统一,并在多个基准测试中取得最佳性能。

摘要:

变化检测(CD)是监测和分析土地覆盖动态变化的基本任务。尽管最近高性能模型和高质量数据集的出现显著推动了该领域的发展,但仍然存在关键限制。当前模型通常只能从单一类型的标注数据中获取有限的知识,无法同时利用多样化的二元变化检测(BCD)和语义变化检测(SCD)数据集。这一限制导致了泛化能力差和适应性有限。最近多模态大语言模型(MLLMs)的进展为统一的CD框架带来了新的可能性。我们利用MLLMs的语言先验和统一能力,开发出UniChange,这是首个基于MLLM的统一变化检测模型。UniChange整合了生成语言能力与专门的CD功能。我们的模型通过引入三个特殊标记:[T1]、[T2]和[CHANGE],成功统一了BCD和SCD任务。此外,UniChange利用文本提示来指导变化类别的识别,消除了对预定义分类头的依赖。这种设计使UniChange能够有效地从多源数据集中获取知识,即使这些数据集的类别定义存在冲突。在四个公开基准(WHU-CD、S2Looking、LEVIR-CD+和SECOND)上的实验表明,UniChange实现了SOTA性能,分别达到了90.41、53.04、78.87和57.62的IoU分数,超过了所有先前方法。代码可在https://github.com/Erxucomeon/UniChange上获取。

英文摘要:

Change detection (CD) is a fundamental task for monitoring and analyzing land cover dynamics. While recent high performance models and high quality datasets have significantly advanced the field, a critical limitation persists. Current models typically acquire limited knowledge from single-type annotated data and cannot concurrently leverage diverse binary change detection (BCD) and semantic change detection (SCD) datasets. This constraint leads to poor generalization and limited versatility. The recent advancements in Multimodal Large Language Models (MLLMs) introduce new possibilities for a unified CD framework. We leverage the language priors and unification capabilities of MLLMs to develop UniChange, the first MLLM-based unified change detection model. UniChange integrates generative language abilities with specialized CD functionalities. Our model successfully unifies both BCD and SCD tasks through the introduction of three special tokens: [T1], [T2], and [CHANGE]. Furthermore, UniChange utilizes text prompts to guide the identification of change categories, eliminating the reliance on predefined classification heads. This design allows UniChange to effectively acquire knowledge from multi-source datasets, even when their class definitions conflict. Experiments on four public benchmarks (WHU-CD, S2Looking, LEVIR-CD+, and SECOND) demonstrate SOTA performance, achieving IoU scores of 90.41, 53.04, 78.87, and 57.62, respectively, surpassing all previous methods. The code is available at https://github.com/Erxucomeon/UniChange.

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。

↑