Commuting Conditional GANS for Multi-Modal Fusion

Siddharth Roheda; Hamid Krim; Benjamin S. Riggan

ICASSP 2020accepted0 citations

Commuting Conditional GANS for Multi-Modal Fusion

Siddharth Roheda, Hamid Krim, Benjamin S. Riggan

Abstract

This paper presents a data driven approach to multi-modal fusion where a hidden latent sub-space between the different modalities is learned. The hidden space is estimated via a bank of Conditional GANs which also commute with each other, leading to an output that lies in a common subspace. Experimental results show improved detection performance compared to existing fusion techniques in ideal as well as noisy sensor condition.

BibTeX

@inproceedings{icassp2020_commutingconditi,
  title = {Commuting Conditional GANS for Multi-Modal Fusion},
  author = {Siddharth Roheda and Hamid Krim and Benjamin S. Riggan},
  booktitle = {ICASSP 2020},
  year = {2020}
}