We track the latest work on Physical-Adversarial-Attack and organize them in the table below in chronological order.
Feel free to submit issues for addition. Thanks! (:heart::heart::heart:)
🔈🔈🔈 Notice: Updated to 2022.12.31⏰
No. |
Title |
Victim Task |
Paper |
Code |
Venue |
1 |
TPatch: A Triggered Physical Adversarial Patch |
Detection |
link |
link |
USENIX Security |
No. |
Title |
Victim Task |
Paper |
Code |
Venue |
1 |
Evaluating the Robustness of Semantic Segmentation for Autonomous Driving against Real-World Adversarial Patch Attacks |
Semantic Segmentation |
link |
link |
WACV |
2 |
Harnessing Perceptual Adversarial Patches for Crowd Counting |
Crowd Counting |
link |
link |
ACM CCS |
3 |
Learning Coated Adversarial Camouflages for Object Detectors |
Vehicle Detection |
link |
--- |
IJCAI |
4 |
FCA: Learning a 3D Full-coverage Vehicle Camouflage for Multi-view Physical Adversarial Attack |
Vehicle Detection |
link |
link |
AAAI |
5 |
Physical Attack on Monocular Depth Estimation with Optimal Adversarial Patches |
Depth Estimation |
link |
--- |
ECCV |
6 |
Shadows can be Dangerous: Stealthy and Effective Physical-world Adversarial Attack by Natural Phenomenon |
Classification |
link |
link |
CVPR |
7 |
Adversarial Color Film: Effective Physical-World Attack to DNNs |
Classification |
link |
--- |
Arxiv |
8 |
DTA: Physical Camouflage Attacks using Differentiable Transformation Network |
Vehicle Detection |
link |
--- |
CVPR |
9 |
Adversarial Texture for Fooling Person Detectors in the Physical World |
Person Detection |
link |
link |
CVPR |
10 |
Infrared Invisible Clothing: Hiding From Infrared Detectors at Multiple Angles in Real World |
Person Detection |
link |
--- |
CVPR |
11 |
Isometric 3D Adversarial Examples in the Physical World |
3D Point Cloud Recognition |
link |
--- |
NIPS |
12 |
Simultaneously Optimizing Perturbations and Positions for Black-box Adversarial Patch Attacks |
Face Recognition |
link |
link |
TPAMI |
No. |
Title |
Victim Task |
Paper |
Code |
Venue |
1 |
Too Good to Be Safe: Tricking Lane Detection in Autonomous Driving with Crafted Perturbations |
Lane detection |
link |
--- |
USENIX Security |
2 |
Invisible Perturbations: Physical Adversarial Examples Exploiting the Rolling Shutter Effect |
Classification |
link |
link |
CVPR |
3 |
Adversarial imaging pipelines |
Classification |
link |
--- |
CVPR |
4 |
Meta-Attack: Class-agnostic and Model-agnostic Physical Adversarial Attack |
Classification |
link |
--- |
ICCV |
5 |
Optical Adversarial Attack |
Classification |
link |
--- |
ICCV |
6 |
Adversarial Laser Beam: Effective Physical-World Attack to DNNs in a Blink |
Classification |
link |
link |
CVPR |
7 |
The Translucent Patch: A Physical and Universal Attack on Object Detectors |
Detection |
link |
--- |
CVPR |
8 |
Naturalistic Physical Adversarial Patch for Object Detectors |
Person Detection |
link |
link |
ICCV |
9 |
Legitimate Adversarial Patches: Evading Human Eyes and Detection Models in the Physical World |
Person Detection |
link |
--- |
ACM MM |
10 |
SLAP: Improving Physical Adversarial Examples with Short-Lived Adversarial Perturbations |
Sign Detection |
link |
link |
USENIX SECURITY |
11 |
Fooling Thermal Infrared Pedestrian Detectors in Real World Using Small Bulbs |
Person Detection |
link |
--- |
AAAI |
12 |
Adv-Makeup: A New Imperceptible and Transferable Attack on Face Recognition |
Face Recognition |
link |
link |
IJCAI |
13 |
Poltergeist: Acoustic Adversarial Machine Learning against Cameras and Computer Vision |
Detection |
link |
link |
IEEE SP |
No. |
Title |
Victim Task |
Paper |
Code |
Venue |
1 |
Bias-based universal adversarial patch attack for automatic check-out |
Classification |
link |
--- |
ECCV |
2 |
Adversarial camouflage: Hiding physical-world attacks with natural styles |
Classification |
link |
link |
CVPR |
3 |
Physgan: Generating physical-world-resilient adversarial examples for autonomous driving |
Classification |
link |
--- |
CVPR |
4 |
Physical adversarial attack on vehicle detector in the carla simulator |
Vehicle Detection |
link |
--- |
Arxiv |
5 |
Adversarial T-shirt! Evading Person Detectors in A Physical World |
Person Detection |
link |
link |
ECCV |
6 |
Universal Physical Camouflage Attacks on Object Detectors |
Detection |
link |
link |
CVPR |
7 |
Making an Invisibility Cloak: Real World Adversarial Attacks on Object Detectors |
Person Detection |
link |
link |
ECCV |
8 |
Adversarial Light Projection Attacks on Face Recognition Systems: A Feasibility Study |
Face Recognition |
link |
--- |
CVPRW |
9 |
Advhat: Real-world adversarial attack on arcface face id system |
Face Recognition |
link |
link |
ICPR |
No. |
Title |
Victim Task |
Paper |
Code |
Venue |
1 |
Attacking Optical Flow |
Optical Flow Estimation |
link |
link |
ICCV |
2 |
Adversarial camera stickers: A physical camera-based attack on deep learning systems |
Classification |
link |
link |
PMLR |
3 |
Perceptual-sensitive gan for generating adversarial patches |
Classification |
link |
--- |
AAAI |
4 |
Fooling automated surveillance cameras: adversarial patches to attack person detection |
Detection |
link |
link |
CVPRW |
5 |
AdvPattern: physical-world attacks on deep person re-identification via adversarially transformable patterns |
Person Re-ID |
link |
link |
ICCV |
6 |
On Adversarial Patches: Real-World Attack on ArcFace-100 Face Recognition System |
Face Recognition |
link |
--- |
SIBIRCON |
No. |
Title |
Victim Task |
Paper |
Code |
Venue |
1 |
Robust Physical-World Attacks on Deep Learning Visual Classification |
Classification |
link |
link |
CVPR |
2 |
Adversarial examples in the physical world |
Classification |
link |
link |
AISS |
3 |
Synthesizing robust adversarial examples |
Classification |
link |
link |
PMLR |
4 |
CAMOU: Learning physical vehicle camouflages to adversarially attack detectors in the wild |
Vehicle Detection |
link |
--- |
ICLR |
5 |
Physical Adversarial Examples for Object Detectors |
Detection |
link |
--- |
USENIX Workshop |
No. |
Title |
Victim Task |
Paper |
Code |
Venue |
1 |
Adversarial Patch |
Classification |
link |
link |
NIPS |
No. |
Title |
Victim Task |
Paper |
Code |
Venue |
1 |
Accessorize to a Crime: Real and Stealthy Attacks on State-of-the-Art Face Recognition |
Face Recognition |
link |
link |
ACM SIGSAC |
⬆ top