DGL-RSIS: Decoupling global spatial context and local class semantics for training-free remote sensing image segmentation
文献类型：期刊论文
| 作者 | Li, Boyi1; Zhang, Ce1; Timmerman, Richard M.1; Bao, Wenxuan2 |
| 刊名 | INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
 |
| 出版日期 | 2026-02-01 |
| 卷号 | 146页码:105113 |
| 关键词 | Vision language model Open-vocabulary semantic segmentation Referring expression segmentation Domain knowledge Training-free |
| ISSN号 | 1569-8432 |
| DOI | 10.1016/j.jag.2026.105113 |
| 产权排序 | 2 |
| 文献子类 | Article |
| 英文摘要 | The emergence of vision language models (VLMs) bridges the gap between vision and language, enabling multimodal understanding beyond traditional visual-only deep learning models. However, transferring VLMs from the natural image domain to remote sensing (RS) segmentation remains challenging due to the large domain gap and the diversity of RS inputs across tasks, particularly in open-vocabulary semantic segmentation (OVSS) and referring expression segmentation (RES). Here, we propose a training-free unified framework, termed DGL-RSIS, which decouples visual and textual representations and performs visual-language alignment at both local semantic and global contextual levels. Specifically, a Global-Local Decoupling (GLD) module decomposes textual inputs into local semantic tokens and global contextual tokens, while image inputs are partitioned into class-agnostic mask proposals. Then, a Local Visual-Textual Alignment (LVTA) module adaptively extracts context-aware visual features from the mask proposals and enriches textual features through knowledgeguided prompt engineering, achieving OVSS from a local perspective. Furthermore, a Global Visual-Textual Alignment (GVTA) module employs a global-enhanced Grad-CAM mechanism to capture contextual cues for referring expressions, followed by a mask selection module that integrates pixel-level activations into mask-level segmentation outputs, thereby achieving RES from a global perspective. Experiments on the iSAID (OVSS) and RRSIS-D (RES) benchmarks demonstrate that DGL-RSIS outperforms existing training-free approaches. Ablation studies further validate the effectiveness of each module. To the best of our knowledge, this is the first unified training-free framework for RS image segmentation, which effectively transfers the semantic capability of VLMs trained on natural images to the RS domain without additional training. |
| URL标识 | 查看原文 |
| WOS研究方向 | Physical Geography ; Remote Sensing |
| 语种 | 英语 |
| WOS记录号 | WOS:001677301100002 |
| 出版者 | ELSEVIER |
| 源URL | [http://ir.igsnrr.ac.cn/handle/311030/220912]  |
| 专题 | 中国科学院地理科学与资源研究所 |
| 通讯作者 | Zhang, Ce |
| 作者单位 | 1.Univ Bristol, Sch Geog Sci, Univ Rd, Bristol BS8 1SS, England; 2.Chinese Acad Sci, Inst Geog Sci & Nat Resources Res, Beijing 100101, Peoples R China |
| 推荐引用方式 GB/T 7714 | Li, Boyi,Zhang, Ce,Timmerman, Richard M.,et al. DGL-RSIS: Decoupling global spatial context and local class semantics for training-free remote sensing image segmentation[J]. INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION,2026,146:105113. |
| APA | Li, Boyi,Zhang, Ce,Timmerman, Richard M.,&Bao, Wenxuan.(2026).DGL-RSIS: Decoupling global spatial context and local class semantics for training-free remote sensing image segmentation.INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION,146,105113. |
| MLA | Li, Boyi,et al."DGL-RSIS: Decoupling global spatial context and local class semantics for training-free remote sensing image segmentation".INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION 146(2026):105113. |
入库方式： OAI收割
来源：地理科学与资源研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。