From Prompts to Self-Prompts: Parameter-Efficient Multi-Label Remote Sensing via Mask-Guided Classification
文献类型：期刊论文
| 作者 | Qu, Ge1; Guan, Xiongwei2; Wen, Fei1; Zou, Xinyu3 |
| 刊名 | REMOTE SENSING
 |
| 出版日期 | 2026-02-05 |
| 卷号 | 18期号:3页码:518 |
| 关键词 | multi-label classification remote sensing self-prompted learning parameter-efficient adaptation foundation models |
| DOI | 10.3390/rs18030518 |
| 产权排序 | 3 |
| 文献子类 | Article |
| 英文摘要 | Multi-label remote sensing scene classification (MLRSSC) requires autonomous discovery of all relevant land-cover categories without human guidance. Conventional expert classifiers return only label vectors without spatial evidence, while foundation segmenters (e.g., SAM, RemoteSAM) remain passively dependent on external prompts-misaligned with autonomous interpretation. We introduce SAFI-XRS, a parameter-efficient self-prompted framework that transforms passive prompting into active scene parsing. By training only <2% of a 332M-parameter segmenter (similar to 2.4M parameters), SAFI-XRS generates class-aligned queries from images via a Semantic Query Generator (SQR), replacing external prompts with self-generated conditioning. A Mask-Guided Classifier (MGC) aggregates spatial evidence into label confidences, enabling mask-based explainability. Experiments on UCM-ML, DFC15-ML, and AID-ML show SAFI-XRS surpasses text-prompted foundation segmenters (+3.9/+3.8 mAP on balanced datasets) while achieving 6.8x parameter efficiency compared to expert models, validating a practical path toward autonomous, explainable RS scene understanding. |
| URL标识 | 查看原文 |
| WOS关键词 | SCENE CLASSIFICATION |
| WOS研究方向 | Environmental Sciences & Ecology ; Geology ; Remote Sensing ; Imaging Science & Photographic Technology |
| 语种 | 英语 |
| WOS记录号 | WOS:001690065800001 |
| 出版者 | MDPI |
| 源URL | [http://ir.igsnrr.ac.cn/handle/311030/220998]  |
| 专题 | 生态系统网络观测与模拟院重点实验室_外文论文 |
| 通讯作者 | Zou, Xinyu |
| 作者单位 | 1.Liaoning Tech Univ, Coll Surveying & Mapping & Geog Sci, Fuxin 123000, Peoples R China; 2.China Univ Geosci Beijing, Sch Artificial Intelligence, Beijing 100083, Peoples R China; 3.Chinese Acad Sci, Key Lab Ecosyst Network Observat & Modeling, Inst Geog Sci & Nat Resources Res, Beijing 100101, Peoples R China |
| 推荐引用方式 GB/T 7714 | Qu, Ge,Guan, Xiongwei,Wen, Fei,et al. From Prompts to Self-Prompts: Parameter-Efficient Multi-Label Remote Sensing via Mask-Guided Classification[J]. REMOTE SENSING,2026,18(3):518. |
| APA | Qu, Ge,Guan, Xiongwei,Wen, Fei,&Zou, Xinyu.(2026).From Prompts to Self-Prompts: Parameter-Efficient Multi-Label Remote Sensing via Mask-Guided Classification.REMOTE SENSING,18(3),518. |
| MLA | Qu, Ge,et al."From Prompts to Self-Prompts: Parameter-Efficient Multi-Label Remote Sensing via Mask-Guided Classification".REMOTE SENSING 18.3(2026):518. |
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。