└── README.md


/README.md:
--------------------------------------------------------------------------------
 1 | # ICCV-2021-adversarial-attacks-and-defense
 2 | ICCV 2021 papers and code focus on adversarial attacks and defense
 3 | 
 4 | ## Attacks
 5 | 
 6 | ### clssification
 7 | * AdvDrop: Adversarial Attack to DNNs by Dropping Information
 8 | * Admix: Enhancing the Transferability of Adversarial Attacks
 9 | * Feature Importance-Aware Transferable Adversarial Attacks
10 | * Consistency-Sensitivity Guided Ensemble Black-Box Adversarial Attacks in Low-Dimensional Spaces
11 | * Augmented Lagrangian Adversarial Attacks
12 |   * [code](https://github.com/jeromerony/augmented_lagrangian_adversarial_attacks)
13 | * LIRA: Learnable, Imperceptible and Robust Backdoor Attacks
14 | * Interpreting Attributions and Interactions of Adversarial Attacks
15 | 
16 | ### detection
17 | 
18 | ### point cloud 
19 | * PointBA: Towards Backdoor Attacks in 3D Point Cloud
20 | * A Backdoor Attack Against 3D Point Cloud Classifiers
21 | 
22 | * Meta Gradient Adversarial Attack
23 | 
24 | ### other tasks
25 | * Adversarial Attack on Deep Cross-Modal Hamming Retrieval `Hamming Retrieval`
26 | 
27 | * Just One Moment: Structural Vulnerability of Deep Action Recognition Against One Frame Attack `Action Recognition`
28 | 
29 | * Parallel Rectangle Flip Attack: A Query-Based Black-Box Attack Against Object Detection `Object Detection`
30 | 
31 | * Practical Relative Order Attack in Deep Ranking `Ranking`
32 | 
33 | * Adversarial Attacks on Multi-Agent Communication
34 | 
35 | * Membership Inference Attacks Are Easier on Difficult Problems `Membership Inference Attacks`
36 | 
37 | * Knowledge-Enriched Distributional Model Inversion Attacks `Model Inversion Attacks`
38 |   * [code]( https://github.com/SCccc21/Knowledge-Enriched-DMI)
39 | * Exploiting Explanations for Model Inversion Attacks `Model Inversion Attacks`
40 | * Aha! Adaptive History-Driven Attack for Decision-Based Black-Box Models
41 | * TkML-AP: Adversarial Attacks to Top-k Multi-Label Learning `Top-k Multi-Label Learning`
42 |   * [code](https://github.com/discovershu/TKML-AP)
43 | 
44 | 
45 | 
46 | * Data-Free Universal Adversarial Perturbation and Black-Box Attack
47 | 
48 | * Attack As the Best Defense: Nullifying Image-to-Image Translation GANs via Limit-Aware Adversarial Attack
49 | 
50 | * Invisible Backdoor Attack With Sample-Specific Triggers
51 | 
52 | * Meta-Attack: Class-Agnostic and Model-Agnostic Physical Adversarial Attack
53 | 
54 | * Attack-Guided Perceptual Data Generation for Real-World Re-Identification
55 | 
56 | * AGKD-BML: Defense Against Adversarial Attack by Attention Guided Knowledge Distillation and Bi-Directional Metric Learning
57 | 
58 | * ProFlip: Targeted Trojan Attack With Progressive Bit Flips
59 | 
60 | 
61 | 
62 | 
63 | ## Defense
64 | 
65 | ### Detection adv
66 | * Multi-Expert Adversarial Attack Detection in Person Re-Identification Using Context Inconsistency
67 | * Black-Box Detection of Backdoor Attacks With Limited Information and Data
68 | * Adversarial Attacks Are Reversible With Natural Supervision
69 | * Rethinking the Backdoor Attacks' Triggers: A Frequency Perspective
70 |   * [code](https://github.com/YiZeng623/frequency-backdoor) 
71 | * Sample Efficient Detection and Classification of Adversarial Attacks via Self-Supervised Embeddings   `exp`
72 | * Detection and Continual Learning of Novel Face Presentation Attacks
73 | * Exploiting Multi-Object Relationships for Detecting Adversarial Attacks in Complex Scenes
74 | 
75 | ### Adv train
76 | * Improving Robustness of Facial Landmark Detection by Defending Against Adversarial Attacks
77 |   * [code]( https://github.com/zhuccly/SAAT)
78 | ## Applications
79 | * Triggering Failures: Out-of-Distribution Detection by Learning From Local Adversarial Attacks in Semantic Segmentation
80 | 
81 | 
82 | 
83 | 
84 | 


--------------------------------------------------------------------------------