Graph-based multimodal fusion with metric learning for multimodal classification

作者：

Highlights：

• Multimodal classification results surpasses the state-of-the-art methods.

• Works with missing modalities.

• Works with any kind and any number of modalities.

摘要

•Multimodal classification results surpasses the state-of-the-art methods.•Works with missing modalities.•Works with any kind and any number of modalities.

论文关键词：Multimodal fusion,Multimodal metric learning,Multimodal classification,Distance graphs

论文评审过程：Received 5 October 2018, Revised 22 March 2019, Accepted 24 June 2019, Available online 28 June 2019, Version of Record 5 July 2019.

论文官网地址：https://doi.org/10.1016/j.patcog.2019.06.013