SIRS: Multitask Joint Learning for Remote Sensing Foreground-Entity Image–Text Retrieval
Published in IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2024
This paper proposes the SIRS cross modal image text retrieval framework, with core contributions including: 1) semantic guided spatial attention, construction of multi task joint learning branches, filtering out noise and refining foreground features; 2) Adaptive multi-scale weighting improves retrieval efficiency. Significant improvement on open-source datasets, with optional output segmentation masks.
Recommended citation:
Z. Zhu, J. Kang, W. Diao, Y. Feng, J. Li and J. Ni, "SIRS: Multitask Joint Learning for Remote Sensing Foreground-Entity Image–Text Retrieval," in IEEE Transactions on Geoscience and Remote Sensing, vol. 62, pp. 1-15, 2024, Art no. 5625615, doi: 10.1109/TGRS.2024.3402216.
Download Paper