Deep learning offers state of the art solutions for image recognition. However, deep models are vulnerable to adversarial perturbations in images that are subtle but significantly change the model's prediction. In a white-box attack, these perturbations are generally learned for deep models that operate on RGB images and, hence, the perturbations are equally distributed in the RGB color space. In this paper, we show that the adversarial perturbations prevail in the Y-channel of the YCbCr space. Our finding is motivated from the fact that the human vision and deep models are more responsive to shape and texture rather than color. Based on our finding, we propose a defense against adversarial images. Our defence, coined ResUpNet, removes perturbations only from the Y-channel by exploiting ResNet features in an upsampling framework without the need for a bottleneck. At the final stage, the untouched CbCr-channels are combined with the refined Y-channel to restore the clean image. Note that ResUpNet is model agnostic as it does not modify the DNN structure. ResUpNet is trained end-to-end in Pytorch and the results are compared to existing defence techniques in the input transformation category. Our results show that our approach achieves the best balance between defence against adversarial attacks such as FGSM, PGD and DDN and maintaining the original accuracies of VGG-16, ResNet50 and DenseNet121 on clean images. We perform another experiment to show that learning adversarial perturbations only for the Y-channel results in higher fooling rates for the same perturbation magnitude.
Deep neural networks are vulnerable to adversarial attacks. The literature is rich with algorithms that can easily craft successful adversarial examples. In contrast, the performance of defense techniques still lags behind. This paper proposes ME-Net, a defense method that leverages matrix estimation (ME). In ME-Net, images are preprocessed using two steps: first pixels are randomly dropped from the image; then, the image is reconstructed using ME. We show that this process destroys the adversarial structure of the noise, while re-enforcing the global structure in the original image. Since hum…
With the advent of deep learning models, face recognition systems have achieved impressive recognition rates. The workhorses behind this success are Convolutional Neural Networks (CNNs) and the availability of large training datasets. However, we show that small human-imperceptible changes to face samples can evade most prevailing face recognition systems. Even more alarming is the fact that the same generator can be extended to other traits in the future. In this work, we present how such a generator can be trained and also extended to other biometric modalities, such as fingerprint recogniti…
National University of Singapore（新加坡国立大学）；Nanjing University（南京大学）
While deep learning in 3D domain has achieved revolutionary performance in many tasks, the robustness of these models has not been sufficiently studied or explored. Regarding the 3D adversarial samples, most existing works focus on manipulation of local points, which may fail to invoke the global geometry properties, like robustness under linear projection that preserves the Euclidean distance, i.e., isometry. In this work, we show that existing state-of-the-art deep 3D models are extremely vulnerable to isometry transformations. Armed with the Thompson Sampling, we develop a black-box attack …
Upon the discovery of adversarial attacks, robust models have become obligatory for deep learning-based systems. Adversarial training with first-order attacks has been one of the most effective defenses against adversarial perturbations to this day. The majority of the adversarial training approaches focus on iteratively perturbing each pixel with the gradient of the loss function with respect to the input image. However, the adversarial training with gradient-based attacks lacks diversity and does not generalize well to natural images and various attacks. This study presents a robust training…
Universidad Autónoma de San Luis Potosí（圣路易斯波托西自治大学）；CICESE Research Center（CICESE研究中心）
In recent years, the security concerns about the vulnerability of Deep Convolutional Neural Networks (DCNN) to Adversarial Attacks (AA) in the form of small modifications to the input image almost invisible to human vision make their predictions untrustworthy. Therefore, it is necessary to provide robustness to adversarial examples in addition to an accurate score when developing a new classifier. In this work, we perform a comparative study of the effects of AA on the complex problem of art media categorization, which involves a sophisticated analysis of features to classify a fine collection…
The trade-off between robustness and accuracy has been widely studied in the adversarial literature. Although still controversial, the prevailing view is that this trade-off is inherent, either empirically or theoretically. Thus, we dig for the origin of this trade-off in adversarial training and find that it may stem from the improperly defined robust error, which imposes an inductive bias of local invariance -- an overcorrection towards smoothness. Given this, we advocate employing local equivariance to describe the ideal behavior of a robust model, leading to a self-consistent robust error …
Differentiable architecture search (DARTS) is a prevailing NAS solution to identify architectures. Based on the continuous relaxation of the architecture space, DARTS learns a differentiable architecture weight and largely reduces the search cost. However, its stability has been challenged for yielding deteriorating architectures as the search proceeds. We find that the precipitous validation loss landscape, which leads to a dramatic performance drop when distilling the final architecture, is an essential factor that causes instability. Based on this observation, we propose a perturbation-base…
Prevailing defense mechanisms against adversarial face images tend to overfit to the adversarial perturbations in the training set and fail to generalize to unseen adversarial attacks. We propose a new self-supervised adversarial defense framework, namely FaceGuard, that can automatically detect, localize, and purify a wide variety of adversarial faces without utilizing pre-computed adversarial training samples. During training, FaceGuard automatically synthesizes challenging and diverse adversarial attacks, enabling a classifier to learn to distinguish them from real faces and a purifier atte…
Face recognition is a prevailing authentication solution in numerous biometric applications. Physical adversarial attacks, as an important surrogate, can identify the weaknesses of face recognition systems and evaluate their robustness before deployed. However, most existing physical attacks are either detectable readily or ineffective against commercial recognition systems. The goal of this work is to develop a more reliable technique that can carry out an end-to-end evaluation of adversarial robustness for commercial systems. It requires that this technique can simultaneously deceive black-b…
Texas A&M University（德克萨斯农工大学）；MIT-IBM Watson AI Lab（麻省理工学院-IBM沃森人工智能实验室）
Neural architecture search (NAS) has witnessed prevailing success in image classification and (very recently) segmentation tasks. In this paper, we present the first preliminary study on introducing the NAS algorithm to generative adversarial networks (GANs), dubbed AutoGAN. The marriage of NAS and GANs faces its unique challenges. We define the search space for the generator architectural variations and use an RNN controller to guide the search, with parameter sharing and dynamic-resetting to accelerate the process. Inception score is adopted as the reward, and a multi-level search strategy i…
Northwestern Polytechnical University（西北工业大学）；University of Cagliari（卡利亚里大学）
Face presentation attack detection (PAD) has become a thorny problem for biometric systems and numerous countermeasures have been proposed to address it. However, majority of them directly extract feature descriptors and distinguish fake faces from the real ones in existing color spaces (e.g. RGB, HSV and YCbCr). Unfortunately, it is unknown for us which color space is the best or how to combine different spaces together. To make matters worse, the real and fake faces are overlapped in existing color spaces. So, in this paper, a learned distinguishable color-liked space is generated to deal wi…
Australian Artificial Intelligence Institute (AAII), University of Technology Sydney（悉尼科技大学澳大利亚人工智能研究所（AAII））；Research School of Computer Science, Australian National University（澳大利亚国立大学计算机科学研究院）；College of Computer Science and Technology, Zhejiang University（浙江大学计算机科学与技术学院）
Most existing works of adversarial samples focus on attacking image recognition models, while little attention is paid to the image retrieval task. In this paper, we identify two inherent challenges in applying prevailing image recognition attack methods to image retrieval. First, image retrieval demands discriminative visual features, which is significantly different from the one-hot class prediction in image recognition. Second, due to the disjoint and potentially unrelated classes between the training and test set in image retrieval, predicting the query category from predefined training cl…
Two major uncertainties, dataset bias and adversarial examples, prevail in state-of-the-art AI algorithms with deep neural networks. In this paper, we present an intuitive explanation for these issues as well as an interpretation of the performance of deep networks in a natural-image space. The explanation consists of two parts: the philosophy of neural networks and a hypothetical model of natural-image spaces. Following the explanation, we 1) demonstrate that the values of training samples differ, 2) provide incremental boost to the accuracy of a CIFAR-10 classifier by introducing an addition…
Prevailing fingerprint recognition systems are vulnerable to spoof attacks. To mitigate these attacks, automated spoof detectors are trained to distinguish a set of live or bona fide fingerprints from a set of known spoof fingerprints. Despite their success, spoof detectors remain vulnerable when exposed to attacks from spoofs made with materials not seen during training of the detector. To alleviate this shortcoming, we approach spoof detection as a one-class classification problem. The goal is to train a spoof detector on only the live fingerprints such that once the concept of "live" has be…
University of Science and Technology of China（中国科学技术大学）；University of Texas at Austin（德克萨斯大学奥斯汀分校）
Deep generative adversarial networks (GANs) have gained growing popularity in numerous scenarios, while usually suffer from high parameter complexities for resource-constrained real-world applications. However, the compression of GANs has less been explored. A few works show that heuristically applying compression techniques normally leads to unsatisfactory results, due to the notorious training instability of GANs. In parallel, the lottery ticket hypothesis shows prevailing success on discriminative models, in locating sparse matching subnetworks capable of training in isolation to full model…
City University of Hong Kong（香港城市大学）；Fudan University（复旦大学）
Data hiding with deep neural networks (DNNs) has experienced impressive successes in recent years. A prevailing scheme is to train an autoencoder, consisting of an encoding network to embed (or transform) secret messages in (or into) a carrier, and a decoding network to extract the hidden messages. This scheme may suffer from several limitations regarding practicability, security, and embedding capacity. In this work, we describe a different computational framework to hide images in deep probabilistic models. Specifically, we use a DNN to model the probability density of cover images, and hide…
