Cross-Modal Retrieval and Semantic Refinement for Remote Sensing Image Captioning

Two-stage remote sensing image captioning (RSIC) methods have achieved promising results by incorporating additional pre-trained remote sensing tasks to extract supplementary information and improve caption quality. However, these methods face limitations in semantic comprehension, as pre-trained de...

Full description

Bibliographic Details

Published in:	Remote Sensing
Main Authors:	Zhengxin Li, Wenzhe Zhao, Xuanyi Du, Guangyao Zhou, Songlin Zhang
Format:	Article
Language:	English
Published:	MDPI AG 2024-01-01
Subjects:	semantic retrieving attention mechanism image captioning remote sensing
Online Access:	https://www.mdpi.com/2072-4292/16/1/196