.png){kind=icon}
.png){kind=icon}
Return
MSSDF: Modality-Shared Self-supervised Distillation for High-Resolution Multi-modal Remote Sensing Image Learning
.png){kind=icon}
PRE
AI
DOI：10.1016/j.inffus.2025.104006.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Proposed a modality-shared self-supervised distillation framework for pertraining. • Designed an information-aware strategy to keep semantic important content. • Introduced a cross-modal mask method to enhance inter-modality correlation learning. • Formulated a multi-task objective for better pretrain task. • Built a 640K 5cm-resolution multi-modal remote sensing dataset.