Towards Robust Training of Multi-Sensor Data Fusion Network Against Adversarial Examples in Semantic Segmentation

The success of multi-sensor data fusions in deep learning appears to be attributed to the use of complementary information among multiple sensor datasets. Compared to their predictive performance, relatively less attention has been devoted to the adversarial robustness of multi-sensor data fusion mo...

Full description

Saved in:
Bibliographic Details
Main Authors: Yu, Youngjoon, Lee, Hong Joo, Kim, Byeong Cheon, Kim, Jung Uk, Ro, Yong Man
Format: Conference Proceeding
Language:English
Subjects:
adversarial examples
adversarial robustness
cumulative learning
Data integration
Deep learning
Knowledge engineering
Multi-sensor data fusions
Predictive models
Semantics
Signal processing
Training
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The success of multi-sensor data fusions in deep learning appears to be attributed to the use of complementary information among multiple sensor datasets. Compared to their predictive performance, relatively less attention has been devoted to the adversarial robustness of multi-sensor data fusion models. To achieve adversarial robust multi-sensor data fusion networks, we propose here a novel robust training scheme called Multi-Sensor Cumulative Learning (MSCL). The motivation behind the MSCL method is based on the way human beings learn new skills. The MSCL allows the multi-sensor fusion network to learn robust features from individual sensors, and then learn complex joint features from multiple sensors just as people learn to walk before they run. The step wise framework of MSCL enables the network to incorporate pre-trained knowledge of robustness with new joint information from multiple sensors. Extensive experimental evidence validated that the MSCL outperforms other multi-sensor fusion training in defending against adversarial examples.
ISSN:2379-190X
DOI:10.1109/ICASSP39728.2021.9413772