Published on Sat Nov 16 2019

SMART: Skeletal Motion Action Recognition aTtack

He Wang, Feixiang He, Zhexi Peng, Yongliang Yang, Tianjia Shao, Kun Zhou, David Hogg

Adversarial attack has inspired great interest in computer vision. We propose a method, SMART, to attack action recognizers which rely on 3D skeletal motions. Our method involves an innovative perceptual loss which ensures the imperceptibility of the attack.

0
0
0
Abstract

Adversarial attack has inspired great interest in computer vision, by showing that classification-based solutions are prone to imperceptible attack in many tasks. In this paper, we propose a method, SMART, to attack action recognizers which rely on 3D skeletal motions. Our method involves an innovative perceptual loss which ensures the imperceptibility of the attack. Empirical studies demonstrate that SMART is effective in both white-box and black-box scenarios. Its generalizability is evidenced on a variety of action recognizers and datasets. Its versatility is shown in different attacking strategies. Its deceitfulness is proven in extensive perceptual studies. Finally, SMART shows that adversarial attack on 3D skeletal motion, one type of time-series data, is significantly different from traditional adversarial attack problems.

Tue Mar 09 2021
Computer Vision
Understanding the Robustness of Skeleton-based Action Recognition under Adversarial Attack
0
0
0
Tue Mar 09 2021
Computer Vision
BASAR:Black-box Attack on Skeletal Action Recognition
Skeletal motion plays a vital role in human activity recognition. The robustness of skeleton-based activity recognizers has been questioned. We propose the first black-box adversarial attack method BASAR.
0
0
0
Thu May 14 2020
Machine Learning
Towards Understanding the Adversarial Vulnerability of Skeleton-based Action Recognition
Skeleton-based action recognition has attracted increasing attention due to its adaptability to dynamic circumstances and potential for broad applications. With the help of deep learning techniques, it has witnessed substantial progress andcurrently achieved around 90\% accuracy in benign environment.
0
0
0
Wed Nov 28 2018
Computer Vision
Adversarial Attacks for Optical Flow-Based Action Recognition Classifiers
Deep learning has catapulted deep models in to production systems that our society is becoming increasingly dependent on. Recent work has shown that largely uninterpretable models exhibit glaring security vulnerabilities. In this work, we develop a powerful untargeted adversarial attack for action recognition systems.
0
0
0
Fri Jan 23 2015
Computer Vision
Advances in Human Action Recognition: A Survey
Human action recognition has been an important topic in computer vision due to its many applications. This survey gives an overview of the most recent advances in human action recognition during the past several years.
0
0
0
Tue Mar 19 2019
Machine Learning
3D human action analysis and recognition through GLAC descriptor on 2D motion and static posture images
This paper presents an approach for identification of actions within depth action videos. First, we process the video to get motion history images(MHIs) and static history images (SHIs) corresponding to an action video. We then characterize the action video by extracting the Gradient Local Auto-
0
0
0