Bullet-Screen-Emoji Attack With Temporal Difference Noise for Video Action Recognition
IEEE Transactions on Circuits and Systems for Video Technology · 2025-01 · 35 (1) · 589-600
DOI: 10.1109/tcsvt.2024.3455799 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Circuits and Systems for Video Technology, 2025-01. Verfasst von Yongkang Zhang und 8 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 87 Literaturangaben. Gefördert von: National Nature Science Foundation of China, Research and Development Program of Beijing Municipal Education Commission, Municipal Natural Science Foundation of Beijing of China.
Abstract (Originalsprache)
Recent studies have shown that video action recognition models are also vulnerable to fooling by adversarial samples. However, currently existing video attack methods usually require high computational overhead (e.g., they generate adversarial perturbations for all frames by default), and most of them are difficult to implement printable attacks in the physical world. To address the above issues, we devise a novel efficient and effective framework for video action recognition attack: Bullet-Screen-Emoji Attack with Temporal Difference Noise (BSE), a reinforcement learning-based black-box attack method that fools the model by simply generating adversarial bullet screens for key frame and scrolling them on clean video. The agent is optimized to make the optimal actions, i.e., searching key frame. Moreover, we introduce a simple and effective temporal difference noise to enhance the attack capability of the adversarial bullet screen and accelerate the convergence speed. Most importantly, BSE enables printable physical attacks. Extensive experiments show that our proposed BSE achieves promising attack performance on mainstream datasets (HMDB51, UCF101 and Kinetics-400) and in the physical world with high efficiency.
Förderung
- National Nature Science Foundation of China
- Research and Development Program of Beijing Municipal Education Commission
- Municipal Natural Science Foundation of Beijing of China
Zitieren
APA
Zhang, Y., Zhang, H., Li, J., Shi, Z., Yang, J., Yang, K., Yin, S., Liang, Q., & Liu, X. (2025). Bullet-Screen-Emoji Attack With Temporal Difference Noise for Video Action Recognition. IEEE Transactions on Circuits and Systems for Video Technology, 35(1), 589-600. https://doi.org/10.1109/tcsvt.2024.3455799
Vancouver
Zhang Y, Zhang H, Li J, Shi Z, Yang J, Yang K, et al. Bullet-Screen-Emoji Attack With Temporal Difference Noise for Video Action Recognition. IEEE Transactions on Circuits and Systems for Video Technology. 2025;35(1):589-600. doi:10.1109/tcsvt.2024.3455799
BibTeX
@article{Zhang2025,
author = {Zhang, Yongkang and Zhang, Han and Li, Jun and Shi, Zhiping and Yang, Jian and Yang, Kaixin and Yin, Shuo and Liang, Qiuyan and Liu, Xianglong},
title = {Bullet-Screen-Emoji Attack With Temporal Difference Noise for Video Action Recognition},
journal = {IEEE Transactions on Circuits and Systems for Video Technology},
year = {2025},
volume = {35},
number = {1},
pages = {589--600},
doi = {10.1109/tcsvt.2024.3455799}
}
EUVIMED