Back to Search Start Over

Cross-Modality Distillation: A case for Conditional Generative Adversarial Networks

Authors :
Roheda, Siddharth
Riggan, Benjamin S.
Krim, Hamid
Dai, Liyi
Publication Year :
2018

Abstract

In this paper, we propose to use a Conditional Generative Adversarial Network (CGAN) for distilling (i.e. transferring) knowledge from sensor data and enhancing low-resolution target detection. In unconstrained surveillance settings, sensor measurements are often noisy, degraded, corrupted, and even missing/absent, thereby presenting a significant problem for multi-modal fusion. We therefore specifically tackle the problem of a missing modality in our attempt to propose an algorithm based on CGANs to generate representative information from the missing modalities when given some other available modalities. Despite modality gaps, we show that one can distill knowledge from one set of modalities to another. Moreover, we demonstrate that it achieves better performance than traditional approaches and recent teacher-student models.<br />Comment: Accepted and Presented at ICASSP 2018

Details

Database :
arXiv
Publication Type :
Report
Accession number :
edsarx.1807.07682
Document Type :
Working Paper