Abstract: The demand for efficient utilization of massive multimodal data has gradually made cross-modal retrieval become a hot topic of concern for many researchers. Based on highly available cross-modal retrieval technology, we can greatly enrich the methods and means of data assurance, which is of great significance in supporting the transformation of command decision-making towards digitization and intelligence. The remote sensing images are taken as the specific data modality, starting with a review of the concept and research frontiers of cross modal retrieval. The typical application scenarios of cross-modal semantic retrieval technology are analyzed, and the main challenges faced by the field development and future research directions are pointed out.