DPATCH: An Adversarial Patch Attack on Object Detectors

By Xin Liu, Huanrui Yang, Ziwei Liu, Linghao Song, Yiran Chen, Hai Helen Li. SafeAI@AAAI, 2018

366 citations recorded by the source corpus.

Object detectors have emerged as an indispensable module in modern computer vision systems. In this work, we propose DPatch -- a black-box adversarial-patch-based attack towards mainstream object detectors (i.e. Faster R-CNN and YOLO). Unlike the original adversarial patch that only manipulates image-level classifier, our DPatch simultaneously attacks the bounding box regression and object classification so as to disable their predictions. Compared to prior works, DPatch has several appealing properties: (1) DPatch can perform both untargeted and targeted effective attacks, degrading the mAP of Faster R-CNN and YOLO from 75.10% and 65.7% down to below 1%, respectively. (2) DPatch is small in size and its attacking effect is location-independent, making it very practical to implement real-world attacks. (3) DPatch demonstrates great transferability among different detectors as well as training datasets. For example, DPatch that is trained on Faster R-CNN can effectively attack YOLO, and vice versa. Extensive evaluations imply that DPatch can perform effective attacks under black-box setup, i.e., even without the knowledge of the attacked network's architectures and parameters. Successful realization of DPatch also illustrates the intrinsic vulnerability of the modern detector architectures to such patch-based adversarial attacks.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Molecular Biology

View Paper: https://www.semanticscholar.org/paper/27d2f8a0601ce4bd73339ddffc83977e858c5503

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

1. Zuxuan Wu
2. Eugene Bagdasarian
3. Vitaly Shmatikov
4. Larry S. Davis
5. Ziwei Liu
6. Shaofeng Li
7. Xinpeng Zhang
8. Haojin Zhu

In Computer Vision and Pattern Recognition

1. Luc Van Gool
2. Ziwei Liu
3. Larry S. Davis
4. Zuxuan Wu
5. Dacheng Tao
6. Ser-Nam Lim
7. Bernt Schiele
8. Ming-Hsuan Yang

In Molecular Biology

1. Xinpeng Zhang
2. Larry S. Davis
3. Vitaly Shmatikov
4. Tom Goldstein
5. Zuxuan Wu
6. Ser-Nam Lim
7. Minghao Guo

Citations in the corpus

• When NAS Meets Robustness: In Search of Robust Architectures Against Adversarial Attacks (2019, background): “Deep neural networks (NNs) can be easily fooled by adversarial examples [9, 32, 20], where effective attacks are proposed such as FGSM [9], BIM [15], C&W [5], DeepFool [22], MI-FGSM [8], and PGD [21].”
• Invisible Backdoor Attacks on Deep Neural Networks Via Steganography and Regularization (2019, background): “4 Network 6CNN+2FC 6CNN+2FC 6CNN+2FC 6CNN+2FC Dataset GTSRB GTSRB GTSRB GTSRB Target classes [1-43] [1-43] [1-43] [1-43] Trigger Number 43 43 43 43 Inject Rate (Each Class) 0.”
• Making an Invisibility Cloak: Real World Adversarial Attacks on Object Detectors (2019, background): “DPatch causes misclassification of detectors, by placing a patch that does not overlap with the objects of interest [22].”
• Blind Backdoors in Deep Learning Models (2020, background): “UAPs require only white-box [8] or black-box [13, 52] access to the model.”