Wei Wang

Wei Wang
Chinese Academy of Sciences | CAS · Institute of Automation

Doctor of Philosophy

About

74
Publications
13,371
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
1,723
Citations
Introduction
Dr. Wei Wang received his PhD degrees in Pattern Recognition from the Institute of Automation, Chinese Academy of Sciences (CASIA) in 2012. He is currently an associate professor of the National Laboratory of Pattern Recognition (NLPR), CASIA. His current research interests include artificial intelligence and its security problem, image and video forensics and steganalysis, and information content security.
Additional affiliations
July 2012 - present
Chinese Academy of Sciences
Position
  • Professor (Assistant)

Publications

Publications (74)
Article
Gait recognition is widely used in social security applications due to its advantages in long-distance human identification. Recently, sequence-based methods have achieved high accuracy by learning abundant temporal and spatial information. However, their robustness under adversarial attacks in an open world has not been clearly explored. In this p...
Preprint
This paper presents the summary report on our DFGC 2022 competition. The DeepFake is rapidly evolving, and realistic face-swaps are becoming more deceptive and difficult to detect. On the contrary, methods for detecting DeepFakes are also improving. There is a two-party game between DeepFake creators and defenders. This competition provides a commo...
Article
Deep neural networks have shown vulnerability to adversarial attacks. Adversarial examples generated with an ensemble of source models can effectively attack unseen target models, posing a security threat to practical applications. In this paper, we investigate the manner of ensemble adversarial attacks from the viewpoint of network gradients with...
Article
Image forgery detection has aroused widespread research interest in both academia and industry because of its potential security threats. Existing forgery detection methods achieve excellent tampered regions localization performance when forged images have not undergone post-processing, which can be detected by observing changes in the statistical...
Chapter
With the identity information in face data more closely related to personal credit and property security, people pay increasing attention to the protection of face data privacy. In different tasks, people have various requirements for face de-identification (De-ID), so we propose a systematical solution compatible for these De-ID operations. Firstl...
Article
Face manipulation techniques improve fast with the development of powerful image generation models. Two particular face manipulation methods, namely face swap and expression reenactment attract much attention for their flexibility and ease to generate high quality synthesis results. Recently, these two subjects are actively studied. However, most e...
Preprint
With the identity information in face data more closely related to personal credit and property security, people pay increasing attention to the protection of face data privacy. In different tasks, people have various requirements for face de-identification (De-ID), so we propose a systematical solution compatible for these De-ID operations. Firstl...
Preprint
This paper presents a summary of the DFGC 2021 competition. DeepFake technology is developing fast, and realistic face-swaps are increasingly deceiving and hard to detect. At the same time, DeepFake detection methods are also improving. There is a two-party game between DeepFake creators and detectors. This competition provides a common platform fo...
Preprint
Deep neural networks have shown their vulnerability to adversarial attacks. In this paper, we focus on sparse adversarial attack based on the $\ell_0$ norm constraint, which can succeed by only modifying a few pixels of an image. Despite a high attack success rate, prior sparse attack methods achieve a low transferability under the black-box protoc...
Preprint
Privacy protection on human biological information has drawn increasing attention in recent years, among which face anonymization plays an importance role. We propose a novel approach which protects identity information of facial images from leakage with slightest modification. Specifically, we disentangle identity representation from other facial...
Article
The development of technologies that can generate Deepfake videos is expanding rapidly. These videos are easily synthesized without leaving obvious traces of manipulation. Though forensically detection in high-definition video datasets has achieved remarkable results, the forensics of compressed videos is worth further exploring. In fact, compresse...
Article
Full-text available
The k-nearest neighbor (KNN) rule is a simple and effective nonparametric classification algorithm in pattern classification. However, it suffers from several problems such as sensitivity to outliers and inaccurate classification decision rule. Thus, a local mean-based k-nearest neighbor classifier (LMKNN) was proposed to address these problems, wh...
Article
Learning to reconstruct 3D shapes using 2D images is an active research topic, with benefits of not requiring expensive 3D data. However, most work in this direction requires multi-view images for each object instance as training supervision, which oftentimes does not apply in practice. In this paper, we relax the common multi-view assumption and e...
Preprint
Full-text available
Images synthesized by powerful generative adversarial network (GAN) based methods have drawn moral and privacy concerns. Although image forensic models have reached great performance in detecting fake images from real ones, these models can be easily fooled with a simple adversarial attack. But, the noise adding adversarial samples are also arousin...
Article
Recent studies highlight the vulnerability of convolutional neural networks (CNNs) to adversarial attacks, which also calls into question the reliability of forensic methods. Existing adversarial attacks generate one-to-one noise, which means these methods have not learned the fingerprint information. Therefore, we introduce two powerful attacks, f...
Preprint
Pose-guided person image generation usually involves using paired source-target images to supervise the training, which significantly increases the data preparation effort and limits the application of the models. To deal with this problem, we propose a novel multi-level statistics transfer model, which disentangles and transfers multi-level appear...
Article
Deep neural networks (DNNs) have seen extensive studies on image recognition and classification, image segmentation, and related topics. However, recent studies show that DNNs are vulnerable in defending adversarial examples. The classification network can be deceived by adding a small amount of perturbation to clean samples. There are challenges w...
Preprint
Learning to reconstruct 3D shapes using 2D images is an active research topic, with benefits of not requiring expensive 3D data. However, most work in this direction requires multi-view images for each object instance as training supervision, which oftentimes does not apply in practice. In this paper, we relax the common multi-view assumption and e...
Preprint
Gait recognition has a broad application in social security due to its advantages in long-distance human identification. Despite the high accuracy of gait recognition systems, their adversarial robustness has not been explored. In this paper, we demonstrate that the state-of-the-art gait recognition model is vulnerable to adversarial attacks. A nov...
Preprint
It is well known that deep learning models are vulnerable to adversarial examples crafted by maliciously adding perturbations to original inputs. There are two types of attacks: targeted attack and non-targeted attack, and most researchers often pay more attention to the targeted adversarial examples. However, targeted attack has a low success rate...
Preprint
Recently, generated images could reach very high quality, even human eyes could not tell them apart from real images. Although there are already some methods for detecting generated images in current forensic community, most of these methods are used to detect a single type of generated images. The new types of generated images are emerging one aft...
Chapter
Recently GAN generated face images are more and more realistic with high-quality, even hard for human eyes to detect. On the other hand, the forensics community keeps on developing methods to detect these generated fake images and try to ensure the credibility of visual contents. Although researchers have developed some methods to detect generated...
Article
The k-nearest neighbor (KNN) algorithm has been widely used in pattern recognition, regression, outlier detection and other data mining areas. However, it suffers from the large distance computation cost, especially when dealing with big data applications. In this paper, we propose a new fast search (FS) algorithm for exact k-nearest neighbors base...
Chapter
Pulse signal is an effective indicator to reflect the physiological and physical state of the human body. There are many heart rate estimation methods in videos and most of them manually design algorithm to modeling noise signal, which is not enough to represent the actual distribution of noise. In this paper, we propose to train a two-layer LSTM t...
Preprint
Recently the GAN generated face images are more and more realistic with high-quality, even hard for human eyes to detect. On the other hand, the forensics community keeps on developing methods to detect these generated fake images and try to guarantee the credibility of visual contents. Although researchers have developed some methods to detect gen...
Chapter
In reversible data embedding, to avoid overflow and underflow problem, before data embedding, boundary pixels are recorded as side information, which may be losslessly compressed. The existing algorithms often assume that a natural image has few boundary pixels so that the size of side information could be rather small. Accordingly, a relatively hi...
Conference Paper
Reversible data hiding (RDH) is a special kind of data hiding technique which can exactly recover the cover image from the stego image after extracting the hidden data. Recently, Wu et al. proposed a novel RDH method with contrast enhancement (RDH-CE). RDH-CE achieved a good effect in improving visual quality especially for poorly illustrated image...
Article
Full-text available
Traditional steganalysis methods usually rely on handcrafted features. However, with the rapid development of advanced steganography, manual design of complex features has become increasingly difficult. In this paper, we propose a new paradigm for steganalysis based on the concept of feature learning. In our method, Convolutional Neural Network (CN...
Preprint
Full-text available
Recently, deep learning has shown its power in steganalysis. However, the proposed deep models have been often learned from pre-calculated noise residuals with fixed high-pass filters rather than from raw images. In this paper, we propose a new end-to-end learning framework that can learn steganalytic features directly from pixels. In the meantime,...
Chapter
Full-text available
In this paper, a novel strategy of Secure Steganography based on Generative Adversarial Networks is proposed to generate suitable and secure covers for steganography. The proposed architecture has one generative network, and two discriminative networks. The generative network mainly evaluates the visual quality of the generated images for steganogr...
Article
Full-text available
A steganographer network corresponds to a graphic structure that the involved vertices (or called nodes) denote social entities such as the data encoders and data decoders, and the associated edges represent any real communicable channels or other social links that could be utilized for steganography. Unlike traditional steganographic algorithms, a...
Article
Steganography aims to conceal the very fact that the communication takes place, by embedding a message into a digit object such as image without introducing noticeable artifacts. A number of steganographic systems have been developed in past years, most of which, however, are confined to the laboratory conditions where the real-world use of stegano...
Conference Paper
Reversible data hiding aims at recovering exactly the cover image from the marked image after extracting the hidden data. Reversible data hiding with contrast enhancement proposed by Wu et al. achieved a good effect in improving visual quality with considerable embedding capacity while PSNR of the marked image is relatively low. In contrast, Predic...
Conference Paper
Image forensics has been focusing on low-level visual features, paying little attention to high-level semantic information of the image. In this work, we propose the framework for image forgery detection based on high-level semantics with three components of image understanding module, the normal rule bank (NR) holding semantic rules that comply wi...
Article
Standing objects on planar surfaces are common to see in images, e.g. people on the ground. For most objects to stay stable on the plane, planar contact is a necessary requirement. However, 2D image splicing usually disregards this physical constraint of 3D world, leading to a potential artifact of object not attached to the plane. This paper is th...
Article
Full-text available
Previous work has shown that feature maps of deep convolutional neural networks (CNNs) can be interpreted as feature representation of a particular image region. Features aggregated from these feature maps have been exploited for image retrieval tasks and achieved state-of-the-art performances in recent years. The key to the success of such methods...
Article
Image forgery is becoming a growing threat to information credibility. Among all kinds of image forgeries, photographic composites of human faces have very serious impacts. To combat this kind of forgery, some forensic methods propose to estimate the 3D lighting environments from different faces and investigate the consistency between them. Althoug...
Article
Most of the quantization based watermarking algorithms are very sensitive to valumetric distortions, while these distortions are regarded as common processing in audio/video analysis. In recent years, watermarking methods which can resist this kind of distortions have attracted a lot of interests. But still many proposed methods can only deal with...
Chapter
Recently, ensemble classifier is predominantly used for steganalysis of digital media, due to its efficiency when working with high-dimensional feature sets and large databases. While fusing the decisions of many weak base classifiers, the majority voting rule is often used, which has the disadvantage that all the classifiers have the same authorit...
Chapter
The key challenge of steganalysis is to construct effective feature representations. Traditional steganalysis systems rely on hand-designed feature extractors. Recently, some efforts have been put toward learning representations automatically using deep models. In this paper, we propose a new CNN based framework for steganalysis based on the concep...
Conference Paper
Image forgery is becoming a growing threat to information credibility. Among all kinds of image forgeries, photographic composites of human faces have very serious impacts. To combat this kind of forgery, some forensic methods propose to estimate the 3D lighting environments from different faces and investigate the consistency between them. Althoug...
Conference Paper
The rapid development in the field of computer graphics (CG) makes it quite easy to create photo-realistic images and videos. This brings forward an emergent requirement for techniques that can distinguish CG from real contents. In this paper, we propose a method that leverages human pulse signal to distinguish between CG and real videos that inclu...
Conference Paper
Correlation of pixels is the most important information used for image steganalysis. Current methods often consider some special types of relationships among neighboring pixels. In this paper, we propose a general descriptor to consider the correlation of pixels comprehensively. We consider the correlation of pixels in an adjacency pattern as a loc...
Conference Paper
Security is an important issue in biometric recognition systems. In recent years, many researchers proposed to use watermarking to improve the security of biometric systems, but some people concern whether the embedded watermarks will influence recognition results. In this paper, we investigate the effects of several fragile and semi-fragile waterm...
Article
Current work on steganalysis for digital images is focused on the construction of complex handcrafted features. This paper proposes a new paradigm for steganalysis to learn features automatically via deep learning models. We novelly propose a customized Convolutional Neural Network for steganalysis. The proposed model can capture the complex depend...
Article
Most of the quantization based watermarking algorithms are very sensitive to valumetric distortions, while these distortions are regarded as common processing in audio/video analysis. In recent years, watermarking methods which can resist this kind of distortions have attracted a lot of interests. But still many proposed methods can only deal with...
Article
In this paper, we focus on local image tampering detection. For a JPEG image, the probability distributions of its DCT coefficients will be disturbed by tampering operation. The tampered region and the unchanged region have different distributions, which is an important clue for locating tampering. Based on the assumption of Laplacian distribution...
Conference Paper
Rational dither modulation (RDM) watermarking was presented to resist amplitude scaling attack. This property is achieved by quantizing the ratio of consecutive samples instead of samples themselves. In this paper, we improve the performance of basic RDM watermarking to resist more types of watermarking attacks. We improve the robustness of our mod...
Conference Paper
In this paper, we focus on detecting data hiding in motion vectors of compressed video and propose a new steganalytic algorithm based on the mutual constraints of motion vectors. The constraints of motion vectors from multiple frames are analyzed and formulized by three functions, then statistical features are extracted based on these functions. Mo...
Conference Paper
Image forensics has now raised the anxiety of justice as increasing cases of abusing tampered images in newspapers and court for evidence are reported recently. With the goal of verifying image content authenticity, passive-blind image tampering detection is called for. More realistic open benchmark databases are also needed to assist the technique...
Article
Full-text available
In this paper, we focus on image tampering detection and tampered region localization. We find that the probability distributions of the DCT coefficients of a JEPG image will be influenced by tampering operation. Hence, we model the distributions of AC DCT coefficients of JPEG image and detect the tampered region from the unchanged region by using...
Conference Paper
Full-text available
With the availability of various digital image edit tools, seeing is no longer believing. In this paper, we focus on tampered region localization for image forensics. We propose an algorithm which can locate tampered region(s) in a lossless compressed tampered image when its unchanged region is output of JPEG decompressor. We find the tampered regi...
Conference Paper
Full-text available
In this paper, we propose a passive image tampering detection method based on modeling edge information. We model the edge image of image chroma component as a finite-state Markov chain and extract low dimensional feature vector from its stationary distribution for tampering detection. The support vector machine (SVM) is utilized as classifier to e...
Conference Paper
Full-text available
A color image splicing detection method based on gray level co-occurrence matrix (GLCM) of thresholded edge image of image chroma is proposed in this paper. Edge images are generated by subtracting horizontal, vertical, main and minor diagonal pixel values from current pixel values respectively and then thresholded with a predefined threshold T. Th...
Conference Paper
In this paper, we investigate our previously developed run-length based features for multi-class blind image steganalysis. We construct a Support Vector Machine classifier for multi-class recognition for both spatial and frequency domain based steganographic algorithms. We also study hierarchical and non-hierarchical multi-class schemes and compare...
Conference Paper
Full-text available
Digital images can be easily tampered with image editing tools. The detection of tampering operations is of great importance. Passive digital image tampering detection aims at verifying the authenticity of digital images without any a prior knowledge on the original images. There are various methods proposed in this filed in recent years. In this...
Conference Paper
Full-text available
In this paper, a simple but efficient approach for blind image splicing detection is proposed. Image splicing is a common and fundamental operation used for image forgery. The detection of image splicing is a preliminary but desirable study for image forensics. Passive detection approaches of image splicing are usually regarded as pattern recogniti...

Network

Cited By

Projects

Project (1)
Project
fake video/image detection