Making an Invisibility Cloak: Real World Adversarial Attacks on Object Detectors
By Zuxuan Wu, Ser-Nam Lim, Larry S. Davis, Tom Goldstein. European Conference on Computer Vision, 2019
325 citations recorded by the source corpus.
We present a systematic study of adversarial attacks on state-of-the-art object detection frameworks. Using standard detection datasets, we train patterns that suppress the objectness scores produced by a range of commonly used detectors, and ensembles of detectors. Through extensive experiments, we benchmark the effectiveness of adversarially trained patches under both white-box and black-box settings, and quantify transferability of attacks between datasets, object classes, and detector models. Finally, we present a detailed study of physical world attacks using printed posters and wearable clothes, and rigorously quantify the performance of such attacks with different metrics.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition
View Paper: https://www.semanticscholar.org/paper/06fa914208bff3b5579ced913241868142daaeff
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
- Mahyar Najibi
- Larry S. Davis
- Ajmal Saeed Mian
- Zheng Xu
- Navid Kardan
- Tom Goldstein
- Naveed Akhtar
In Computer Vision and Pattern Recognition
- Ajmal Saeed Mian
- Mubarak Shah
- Luc Van Gool
- Larry S. Davis
- Dacheng Tao
- Jitendra Malik
- Xiaogang Wang
- Bernt Schiele
Citations in the corpus
- Advances in adversarial attacks and defenses in computer vision: A survey (2021, background): “Another example of deceiving object detector in the real-world can be found in [210].”
- Universal Adversarial Training (2018, background): “Many works have shown that these vulnerabilities can be exploited by showing real world attacks on face detection (Sharif et al., 2016), object detection (Wu et al., 2019), and copyright detection (Saadatpanah, Shafahi, and Goldstein, 2019).”
- Fawkes: Protecting Privacy against Unauthorized Deep Learning Models (2020, background): “Others produce adversarial patches in the form of bright patterns printed on sweatshirts or signs, which prevent facial recognition algorithms from even registering their wearer as a person [54, 64].”