Just One Moment: Structural Vulnerability of Deep Action Recognition against One Frame Attack

2020/11/30 by Jaehui Hwang, Hwang, Jaehui, Junhyuk Kim +5 · 1 citation
Computer Science · Medicine · #Adversarial Robustness in Machine Learning #Anomaly Detection Techniques and Applications #Cardiac Arrest and Resuscitation #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Machine Learning (cs.LG)

paper · pdf · doi:10.48550/arxiv.2011.14585

openalex publication_date 2020/11/30 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

The video-based action recognition task has been extensively studied in recent years. In this paper, we study the structural vulnerability of deep learning-based action recognition models against the adversarial attack using the one frame attack that adds an inconspicuous perturbation to only a single frame of a given video clip. Our analysis shows that the models are highly vulnerable against the one frame attack due to their structural properties. Experiments demonstrate high fooling rates and inconspicuous characteristics of the attack. Furthermore, we show that strong universal one frame perturbations can be obtained under various scenarios. Our work raises the serious issue of adversarial vulnerability of the state-of-the-art action recognition models in various perspectives.

Cited by

• Robustness Evaluation for Video Models with Reinforcement Learning

Related

• Multimodal Attack Detection for Action Recognition Models
• Appending Adversarial Frames for Universal Video Attack
• Adversarial Attacks for Optical Flow-Based Action Recognition Classifiers
• Inter-frame Accelerate Attack against Video Interpolation Models
• Understanding the Robustness of Skeleton-based Action Recognition under Adversarial Attack