AdvHat: Real-World Adversarial Attack on ArcFace Face ID System
By Stepan Alekseevich Komkov, Aleksandr Petiushko. International Conference on Pattern Recognition, 2019
362 citations recorded by the source corpus.
In this paper we propose a novel easily reproducible technique to attack the best public Face ID system ArcFace in different shooting conditions. To create an attack, we print the rectangular paper sticker on a common color printer and put it on the hat. The adversarial sticker is prepared with a novel algorithm for off-plane transformations of the image which imitates sticker location on the hat. Such an approach confuses the state-of-the-art public Face ID model LResNet100E-IR, ArcFace@ms1m-refine-v2 and is transferable to other Face ID models.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Signal Processing
View Paper: https://www.semanticscholar.org/paper/e3dacba4399fc66f31779060db730ea423f341d8
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Mubarak Shah
- Zuxuan Wu
- Ajmal Saeed Mian
- Larry S. Davis
- Navid Kardan
- Naveed Akhtar
- Ser-Nam Lim
- Tom Goldstein
In Computer Vision and Pattern Recognition
- Luc Van Gool
- Mubarak Shah
- Ajmal Saeed Mian
- Dacheng Tao
- Larry S. Davis
- Xiaogang Wang
- Ming-Hsuan Yang
- Zuxuan Wu
In Signal Processing
- Mubarak Shah
- Navid Kardan
- Naveed Akhtar
- Ajmal Saeed Mian
- Emily Wenger
- Ben Y. Zhao
- Baofu Wu
- Larry S. Davis
Citations in the corpus
- Fawkes: Protecting Privacy against Unauthorized Deep Learning Models (2020, background): “Komkov and Petiushko [23] showed that carefully computed adversarial stickers on a hat can reduce its wearer’s likelihood of being recognized.”
- Advances in adversarial attacks and defenses in computer vision: A survey (2021, background): “[253] S. Komkov and A. Petiushko, ‘‘AdvHat: Real-world adversarial attack on ArcFace face ID system,’’ 2019, arXiv:1908.08705.”
- Making an Invisibility Cloak: Real World Adversarial Attacks on Object Detectors (2019, background): “generate stickers attached to hats to attack face classifiers [11].”
- Computing Systems for Autonomous Driving: State of the Art and Challenges (2020, background): “For example, a photograph instead of a human to pass the face recognition or a note-sized photograph posted on the forehead makes machine learning algorithms fail to detect faces [232].”
- Deep Learning Sensor Fusion for Autonomous Vehicle Perception and Localization: A Review (2020, neutral): “AdvHat introduced in [177] is an interesting adversarial attack method that attacks face ID systems.”