Can't Slow me Down: Learning Robust and Hardware-Adaptive Object Detectors against Latency Attacks for Edge Devices

TL;DR

This paper proposes a hardware-aware adversarial training method to defend object detectors against latency attacks, significantly improving real-time processing speed and robustness on edge devices.

Contribution

It introduces a background-attentive adversarial training approach that considers hardware capabilities to defend against latency attacks in object detection.

Findings

Restores processing speed from 13 FPS to 43 FPS on Jetson Orin NX.

Achieves a better balance between clean and robust accuracy.

Demonstrates effectiveness of the proposed defense through extensive experiments.

Abstract

Object detection is a fundamental enabler for many real-time downstream applications such as autonomous driving, augmented reality and supply chain management. However, the algorithmic backbone of neural networks is brittle to imperceptible perturbations in the system inputs, which were generally known as misclassifying attacks. By targeting the real-time processing capability, a new class of latency attacks are reported recently. They exploit new attack surfaces in object detectors by creating a computational bottleneck in the post-processing module, that leads to cascading failure and puts the real-time downstream tasks at risks. In this work, we take an initial attempt to defend against this attack via background-attentive adversarial training that is also cognizant of the underlying hardware capabilities. We first draw system-level connections between latency attack and hardware capacity across heterogeneous GPU devices. Based on the particular adversarial behaviors, we utilize objectness loss as a proxy and build background attention into the adversarial training pipeline, and achieve a reasonable balance between clean and robust accuracy. The extensive experiments demonstrate the defense effectiveness of restoring real-time processing capability from $13$ FPS to $43$ FPS on Jetson Orin NX, with a better trade-off between the clean and robust accuracy.