cs.CVNov 12, 2026

Adversarial Tracking: Attacks on Image Sensors for Object Tracking in Autonomous Vehicles

Authors: Han Wu, Alex Chan, Ekin Can Erkus, Tousif Rahman, Rishad Shafik

Organizations: School of Engineering, Newcastle University, Newcastle upon Tyne, United Kingdom

Abstract

Object tracking systems estimate the motion of surrounding vehicles by linking detected objects across video frames. In particular, Tracking-By-Detection (TBD) systems combine object detectors with motion prediction and data association to maintain object identities over time. However, neural network-based detectors are vulnerable to adversarial attacks that perturb input images, and their impact on downstream trajectory estimation is not fully characterized by detection accuracy alone. Therefore, this paper evaluates image-specific and image-agnostic attacks on YOLOv3 and YOLOv4 detectors paired with SORT, Deep SORT, Strong SORT, and OC-SORT tracking systems. The evaluation uses CARLA simulator and measures tracking quality through Multiple Object Tracking Accuracy (MOTA), Higher Order Tracking Accuracy (HOTA), and Identification F1 (IDF1). Under image-specific attacks, HOTA for YOLOv3-based trackers declines from 53.45-56.19% to 14.85-17.03%, while MOTA reaches -772.60% in the worst case. YOLOv4-based trackers exhibit greater robustness, while HOTA still declines to 25.06-26.86%. In addition, image-agnostic attacks can reduce the tracking accuracy using a single Universal Adversarial Perturbation (UAP), without generating a separate perturbation for individual frames.

Figures & tables

Explore similar work

Apr 1, 2026

Out of Sight, Out of Track: Adversarial Attacks on Propagation-based Multi-Object Trackers via Query State Manipulation

Recent Tracking-by-Query-Propagation (TBP) methods have advanced Multi-Object Tracking (MOT) by enabling end-to-end (E2E) pipelines with long-range temporal modeling. However, this reliance on query propagation introduces unexplored architectural vulnerabilities to adversarial attacks. We present FADE, a novel attack framework designed to exploit these specific vulnerabilities. FADE employs two attack strategies targeting core TBP mechanisms: (i) Temporal Query Flooding: Generates spurious temporally consistent track queries to exhaust the tracker's limited query budget, forcing it to terminate valid tracks. (ii) Temporal Memory Corruption: Directly attacks the query updater's memory by severing temporal links via state de-correlation and erasing the learned feature identity of matched tracks. Furthermore, we introduce a differentiable pipeline to optimize these attacks for physical-world realizability by leveraging simulations of advanced perception sensor spoofing. Experiments on MOT17 and MOT20 benchmarks demonstrate that FADE is highly effective against state-of-the-art TBP trackers, causing significant identity switches and track terminations.
Oct 27, 2026

Detector-Integrated Multi-Cue Association for Occlusion-Aware UAV Multi-Object Tracking

Multi-object tracking (MOT) in unmanned aerial vehicle (UAV) imagery is challenging because vehicles are small, visually similar, and frequently occluded, which leads to identity switches (IDSW). Appearance-based trackers such as DeepSORT rely on a standalone re-identification (ReID) network, whose per-target inference limits runtime speed, while the appearance of small or occluded vehicles remains unreliable. This paper proposes a detector-integrated multi-cue tracking framework. Lightweight projection heads are attached to the neck of the YOLOv8n detector to extract target embeddings through ROIAlign, producing detections and embeddings in a single forward pass. A multi-cue association cost then combines motion, IoU, and appearance distances with weights determined through ablation, complemented by low-score rescue and appearance-memory protection. On the UAVDT benchmark, the proposed method achieves a MOTA of 28.33, a HOTA of 31.16, and an IDF1 of 38.18 at 82.59 FPS, the best among the compared trackers in all three metrics, and reduces IDSW by 58.4% compared with DeepSORT.
Aug 17, 2026

AdROD: HyperNetwork-based Adversarially Robust Object Detection for Autonomous Driving

Camera-based object detectors are vulnerable to physical adversarial attacks designed to suppress detections. While adversarial training and input purification offer some protection, they often overfit to specific attack distributions and fail on adaptive adversaries. This paper presents AdROD, an embedded, stochastic ensemble defense software designed for autonomous driving. AdROD employs {\em low-rank HyperNetworks}, which require only 1.6% of the parameter footprint of standard HyperNetworks, to generate diverse detectors at a per-frame rate, making it impractical for attackers to obtain the deployed detectors in time. To further improve adversarial robustness, AdROD incorporates a novel functional diversity mechanism, which couples stochastic weight updates with unique input-space transformations. We design two serving modes of AdROD that strike different trade-offs between robustness and runtime overhead: AdROD-I, a continuous protection mode for maximum resilience that leverages inter-detector disagreement to recover compromised detections, and AdROD-II, an on-demand mode triggered by kinematic discontinuities in object tracking. Through comprehensive evaluation with synthetic benchmarks, physically deployed adversarial patches, and end-to-end safety tests in the OpenCDA co-simulator, AdROD outperforms five baseline defenses and exhibits superior generalizability compared with the evaluated adversarial-training baselines, while maintaining real-time performance for safely stopping the vehicle at a stop sign instrumented with adversarial patches.