Boosting Adversarial Transferability for Skeleton-based Action Recognition via Exploring the Model Posterior Space

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Diao, Yunfeng, Wu, Baiqi, Zhang, Ruixuan, Yang, Xun, Wang, Meng, Wang, He
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910589923622912
author Diao, Yunfeng
Wu, Baiqi
Zhang, Ruixuan
Yang, Xun
Wang, Meng
Wang, He
author_facet Diao, Yunfeng
Wu, Baiqi
Zhang, Ruixuan
Yang, Xun
Wang, Meng
Wang, He
contents Skeletal motion plays a pivotal role in human activity recognition (HAR). Recently, attack methods have been proposed to identify the universal vulnerability of skeleton-based HAR(S-HAR). However, the research of adversarial transferability on S-HAR is largely missing. More importantly, existing attacks all struggle in transfer across unknown S-HAR models. We observed that the key reason is that the loss landscape of the action recognizers is rugged and sharp. Given the established correlation in prior studies~\cite{qin2022boosting,wu2020towards} between loss landscape and adversarial transferability, we assume and empirically validate that smoothing the loss landscape could potentially improve adversarial transferability on S-HAR. This is achieved by proposing a new post-train Dual Bayesian strategy, which can effectively explore the model posterior space for a collection of surrogates without the need for re-training. Furthermore, to craft adversarial examples along the motion manifold, we incorporate the attack gradient with information of the motion dynamics in a Bayesian manner. Evaluated on benchmark datasets, e.g. HDM05 and NTU 60, the average transfer success rate can reach as high as 35.9\% and 45.5\% respectively. In comparison, current state-of-the-art skeletal attacks achieve only 3.6\% and 9.8\%. The high adversarial transferability remains consistent across various surrogate, victim, and even defense models. Through a comprehensive analysis of the results, we provide insights on what surrogates are more likely to exhibit transferability, to shed light on future research.
format Preprint
id arxiv_https___arxiv_org_abs_2407_08572
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Boosting Adversarial Transferability for Skeleton-based Action Recognition via Exploring the Model Posterior Space
Diao, Yunfeng
Wu, Baiqi
Zhang, Ruixuan
Yang, Xun
Wang, Meng
Wang, He
Computer Vision and Pattern Recognition
Skeletal motion plays a pivotal role in human activity recognition (HAR). Recently, attack methods have been proposed to identify the universal vulnerability of skeleton-based HAR(S-HAR). However, the research of adversarial transferability on S-HAR is largely missing. More importantly, existing attacks all struggle in transfer across unknown S-HAR models. We observed that the key reason is that the loss landscape of the action recognizers is rugged and sharp. Given the established correlation in prior studies~\cite{qin2022boosting,wu2020towards} between loss landscape and adversarial transferability, we assume and empirically validate that smoothing the loss landscape could potentially improve adversarial transferability on S-HAR. This is achieved by proposing a new post-train Dual Bayesian strategy, which can effectively explore the model posterior space for a collection of surrogates without the need for re-training. Furthermore, to craft adversarial examples along the motion manifold, we incorporate the attack gradient with information of the motion dynamics in a Bayesian manner. Evaluated on benchmark datasets, e.g. HDM05 and NTU 60, the average transfer success rate can reach as high as 35.9\% and 45.5\% respectively. In comparison, current state-of-the-art skeletal attacks achieve only 3.6\% and 9.8\%. The high adversarial transferability remains consistent across various surrogate, victim, and even defense models. Through a comprehensive analysis of the results, we provide insights on what surrogates are more likely to exhibit transferability, to shed light on future research.
title Boosting Adversarial Transferability for Skeleton-based Action Recognition via Exploring the Model Posterior Space
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2407.08572