paper-with-me

Papers

A Generative Victim Model for Segmentation

2024-12-10 · Aixuan Li, Jing Zhang, Jiawei Shi, Yiran Zhong, Yuchao Dai

We find that the well-trained victim models (VMs), against which the attacks are generated, serve as fundamental prerequisites for adversarial attacks, i.e. a segmentation VM is needed to generate attacks for segmentation. In this context, the victim model is assumed to be robust to achieve effective adversarial perturbation generation. Instead of focusing on improving the robustness of the task-specific victim models, we shift our attention to image generation. From an image generation perspective, we derive a novel VM for segmentation, aiming to generate adversarial perturbations for segmentation tasks without requiring models explicitly designed for image segmentation. Our approach to adversarial attack generation diverges from conventional white-box or black-box attacks, offering a fresh outlook on adversarial attack strategies. Experiments show that our attack method is able to generate effective adversarial attacks with good transferability.

📄 PDF Abstract BibTeX arXiv:2412.07274

Code (0)

등록된 구현이 없습니다.

Tasks

Adversarial AttackImage GenerationImage SegmentationmodelSegmentationSemantic Segmentation

Methods 이 논문이 사용한 방법론

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Attention 설명 없음

Similar Papers 제목 키워드 기반

ConSeg: Contextual Backdoor Attack Against Semantic Segmentation

2025-07-26 · Bilal Hussain Abbasi, Zirui Gong, Yanjun Zhang, Shang Gao 외 arxiv

Despite significant advancements in computer vision, semantic segmentation models may be susceptible to backdoor attacks. These attacks, involving hidden triggers, aim to cause the models to misclassify instances of the …

Semantic Segmentation

Influencer Backdoor Attack on Semantic Segmentation

2023-03-21 · Haoheng Lan, Jindong Gu, Philip Torr, Hengshuang Zhao

When a small number of poisoned samples are injected into the training dataset of a deep neural network, the network can be induced to exhibit malicious behavior during inferences, which poses potential threats to real-w…

Backdoor AttackPositionSegmentationSemantic Segmentation

Ensemble-based Blackbox Attacks on Dense Prediction

2023-03-25 · CVPR 2023 1 · Zikui Cai, Yaoteng Tan, M. Salman Asif

We propose an approach for adversarial attacks on dense prediction models (such as object detectors and segmentation). It is well known that the attacks generated by a single surrogate model do not transfer to arbitrary …

Objectobject-detectionObject DetectionPrediction+1

ZeroPur: Succinct Training-Free Adversarial Purification

2024-06-05 · Erhu Liu, Zonglin Yang, Bo Liu, Bin Xiao 외

Adversarial purification is a kind of defense technique that can defend against various unseen adversarial attacks without modifying the victim classifier. Existing methods often depend on external generative models or c…

Adversarial Purification

GAMA: Generative Adversarial Multi-Object Scene Attacks

2022-09-20 · Abhishek Aich, Calvin-Khang Ta, Akash Gupta, Chengyu Song 외

The majority of methods for crafting adversarial attacks have focused on scenes with a single dominant object (e.g., images from ImageNet). On the other hand, natural scenes include multiple dominant objects that are sem…

Language ModellingObject