BASAR:Black-box Attack on Skeletal Action Recognition

Abstract

Skeletal motion plays a vital role in human activity recognition as either an independent data source or a complement. The robustness of skeleton-based activity recognizers has been questioned recently, which shows that they are vulnerable to adversarial attacks when the full-knowledge of the recognizer is accessible to the attacker. However, this white-box requirement is overly restrictive in most scenarios and the attack is not truly threatening. In this paper, we show that such threats do exist under black-box settings too. To this end, we propose the first black-box adversarial attack method BASAR. Through BASAR, we show that adversarial attack is not only truly a threat but also can be extremely deceitful, because on-manifold adversarial samples are rather common in skeletal motions, in contrast to the common belief that adversarial samples only exist off-manifold. Through exhaustive evaluation and comparison, we show that BASAR can deliver successful attacks across models, data, and attack modes. Through harsh perceptual studies, we show that it achieves effective yet imperceptible attacks. By analyzing the attack on different activity recognizers, BASAR helps identify the potential causes of their vulnerability and provides insights on what classifiers are likely to be more robust against attack. Code is available at https://github.com/realcrane/BASAR-Black-box-Attack-on-Skeletal-Action-Recognition.

arXiv commentAccepted in CVPR 2021

References in corpus (16)

• Intriguing properties of neural networks
• The Kinetics Human Action Video Dataset
• ZOO: Zeroth Order Optimization based Black-box Attacks to Deep Neural Networks without Training Substitute Models
• Transferability in Machine Learning: from Phenomena to Black-Box Attacks using Adversarial Samples
• An End-to-End Spatio-Temporal Attention Model for Human Action Recognition from Skeleton Data
• Adversarial Attacks and Defences: A Survey
• The Space of Transferable Adversarial Examples
• Adversarial Attack and Defense on Graph Data: A Survey
• Adversarial Transformation Networks: Learning to Generate Adversarial Examples
• Adversarial Attacks on Deep Neural Networks for Time Series Classification
• A Survey on 3D Skeleton-Based Action Recognition Using Learning Method
• Adversarial Attacks for Optical Flow-Based Action Recognition Classifiers
• Towards Understanding the Adversarial Vulnerability of Skeleton-based Action Recognition
• Understanding the Robustness of Skeleton-based Action Recognition under Adversarial Attack
• SMART: Skeletal Motion Action Recognition aTtack
• Adversarial Attacks on Multivariate Time Series