In this paper, we propose IVMEA, a multi-modal EA framework under imbalanced visual modality information. Specifically, IVMEA first establishes a mapping network from semantic features to image ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback