88 citations · 199 across the 18 of their papers we have counts for
20 papers
VT-3DAD: Cross-Category 3D Anomaly Detection via Visual-Text Normal Space Alignment
Zi Wang, Katsuya Hotta, Yawen Zou +4
Few-shot cross-category 3D anomaly detection aims to determine whether an unknown point cloud belongs to a target normal category using only a few normal references. Existing train…
Investigating Top- White-Box and Transferable Black-box Attack
Chaoning Zhang, Philipp Benz, Adil Karjauv +3
Existing works have identified the limitation of top- attack success rate (ASR) as a metric to evaluate the attack strength but exclusively investigated it in the white-box sett…
Dual Temperature Helps Contrastive Learning Without Many Negative Samples: Towards Understanding and Simplifying MoCo
Chaoning Zhang, Kang Zhang, Trung X. Pham +4
Contrastive learning (CL) is widely known to require many negative samples, 65536 in MoCo for instance, for which the performance of a dictionary-free framework is often inferior b…
How Does SimSiam Avoid Collapse Without Negative Samples? A Unified Understanding with Self-supervised Contrastive Learning
Chaoning Zhang, Kang Zhang, Chenshuang Zhang +3
To avoid collapse in self-supervised learning (SSL), a contrastive loss is widely used but often requires a large number of negative samples. Without negative samples yet achieving…
Towards Transferable Unrestricted Adversarial Examples with Minimum Changes
Fangcheng Liu, Chao Zhang, Hongyang Zhang
Transfer-based adversarial example is one of the most important classes of black-box attacks. However, there is a trade-off between transferability and imperceptibility of the adve…
Unrestricted Adversarial Attacks on ImageNet Competition
Yuefeng Chen, Xiaofeng Mao, Yuan He +34
Many works have investigated the adversarial attacks or defenses under the settings where a bounded and imperceptible perturbation can be added to the input. However in the real-wo…
Restoration of Video Frames from a Single Blurred Image with Motion Understanding
Dawit Mureja Argaw, Junsik Kim, Francois Rameau +2
We propose a novel framework to generate clean video frames from a single motion-blurred image. While a broad range of literature focuses on recovering a single image from a blurre…
Universal Adversarial Training with Class-Wise Perturbations
Philipp Benz, Chaoning Zhang, Adil Karjauv +1
Despite their overwhelming success on a wide range of applications, convolutional neural networks (CNNs) are widely recognized to be vulnerable to adversarial examples. This intrig…
Universal Adversarial Perturbations Through the Lens of Deep Steganography: Towards A Fourier Perspective
Chaoning Zhang, Philipp Benz, Adil Karjauv +1
The booming interest in adversarial attacks stems from a misalignment between human vision and a deep neural network (DNN), i.e. a human imperceptible perturbation fools the DNN. M…
Self-Adaptive Training: Bridging Supervised and Self-Supervised Learning
Lang Huang, Chao Zhang, Hongyang Zhang
We propose self-adaptive training -- a unified training algorithm that dynamically calibrates and enhances training processes by model predictions without incurring an extra comput…
Towards Robust Data Hiding Against (JPEG) Compression: A Pseudo-Differentiable Deep Learning Approach
Chaoning Zhang, Adil Karjauv, Philipp Benz +1
Data hiding is one widely used approach for protecting authentication and ownership. Most multimedia content like images and videos are transmitted or saved in the compressed form.…
ResNet or DenseNet? Introducing Dense Shortcuts to ResNet
Chaoning Zhang, Philipp Benz, Dawit Mureja Argaw +5
ResNet or DenseNet? Nowadays, most deep learning based approaches are implemented with seminal backbone networks, among them the two arguably most famous ones are ResNet and DenseN…
CD-UAP: Class Discriminative Universal Adversarial Perturbation
Chaoning Zhang, Philipp Benz, Tooba Imtiaz +1
A single universal adversarial perturbation (UAP) can be added to all natural images to change most of their predicted class labels. It is of high practical relevance for an attack…
Double Targeted Universal Adversarial Perturbations
Philipp Benz, Chaoning Zhang, Tooba Imtiaz +1
Despite their impressive performance, deep neural networks (DNNs) are widely known to be vulnerable to adversarial attacks, which makes it challenging for them to be deployed in se…
Robustness May Be at Odds with Fairness: An Empirical Study on Class-wise Accuracy
Philipp Benz, Chaoning Zhang, Adil Karjauv +1
Convolutional neural networks (CNNs) have made significant advancement, however, they are widely known to be vulnerable to adversarial attacks. Adversarial training is the most wid…
Revisiting Batch Normalization for Improving Corruption Robustness
Philipp Benz, Chaoning Zhang, Adil Karjauv +1
The performance of DNNs trained on clean images has been shown to decrease when the test images have common corruptions. In this work, we interpret corruption robustness as a domai…
Batch Normalization Increases Adversarial Vulnerability and Decreases Adversarial Transferability: A Non-Robust Feature Perspective
Philipp Benz, Chaoning Zhang, In So Kweon
Batch normalization (BN) has been widely used in modern deep neural networks (DNNs) due to improved convergence. BN is observed to increase the model accuracy while at the cost of…
Data from Model: Extracting Data from Non-robust and Robust Models
Philipp Benz, Chaoning Zhang, Tooba Imtiaz +1
The essence of deep learning is to exploit data to train a deep neural network (DNN) model. This work explores the reverse process of generating data from a model, attempting to re…
Understanding Adversarial Examples from the Mutual Influence of Images and Perturbations
Chaoning Zhang, Philipp Benz, Tooba Imtiaz +1
A wide variety of works have explored the reason for the existence of adversarial examples, but there is no consensus on the explanation. We propose to treat the DNN logits as a ve…
Self-Adaptive Training: beyond Empirical Risk Minimization
Lang Huang, Chao Zhang, Hongyang Zhang
We propose self-adaptive training---a new training algorithm that dynamically corrects problematic training labels by model predictions without incurring extra computational cost--…