Published April 21, 2021 | Version v1
Conference paper Open

DriveGuard: Robustification of Automated Driving Systems with Deep Spatio-Temporal Convolutional Autoencoder

  • 1. KIOS Center of Excellence, University of Cyprus

Description

Autonomous vehicles increasingly rely on cameras to provide the input for perception and scene understanding and the ability of these models to classify their environment and objects, under adverse conditions and image noise is crucial. When the input is, either unintentionally or through targeted attacks, deteriorated, the reliability of autonomous vehicle is compromised. In order to mitigate such phenomena, we propose DriveGuard, a lightweight spatio-temporal autoencoder, as a solution to robustify the image segmentation process for autonomous vehicles. By first processing camera images with DriveGuard, we offer a more universal solution than having to re-train each perception model with noisy input. We explore the space of different autoencoder architectures and evaluate them on a diverse dataset created with real and synthetic images demonstrating that by exploiting spatio-temporal information combined with multi-component loss we significantly increase robustness against adverse image effects reaching within 5-6% of that of the original model on clean images.

Notes

© 2021IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. https://www.ieee.org/publications_standards/publications/rights/rights_policies.html A. Papachristodoulou, C. Kyrkou and T. Theocharides, "DriveGuard: Robustification of Automated Driving Systems with Deep Spatio-Temporal Convolutional Autoencoder," 2021 IEEE Winter Conference on Applications of Computer Vision Workshops (WACVW), 2021, pp. 107-116, doi: 10.1109/WACVW52041.2021.00016. https://ieeexplore.ieee.org/document/9407818

Files

DriveGuard.pdf

Files (3.9 MB)

Name Size Download all
md5:78cde4c71fab9213f8b845da8529e3ad
3.9 MB Preview Download

Additional details

Funding

CARAMEL – Artificial Intelligence based cybersecurity for connected and automated vehicles 833611
European Commission
KIOS CoE – KIOS Research and Innovation Centre of Excellence 739551
European Commission