Fuhao Zou

Real-Time Compressed- Domain Video Watermarking Resistance to Geometric Distortions

A proposed real-time video watermarking scheme is transparent and robust to geometric distortions, including rotation with cropping, scaling, aspect ratio change, frame dropping, and swapping.

Efficient Image Copy Detection Using Multiscale Fingerprints

A multiscale scale-invariant feature transform (SIFT) descriptor can help improve our ability to discriminate between images when using copy detection to identify illegal image copies.

Scalable Multimedia Retrieval by Deep Learning Hashing with Relative Similarity Learning

Learning-based hashing methods are becoming the mainstream for approximate scalable multimedia retrieval. They consist of two main components: hash codes learning for training data and hash functions learning for new data points. Tremendous efforts have been devoted to designing novel methods for these two components, i.e., supervised and unsupervised methods for learning hash codes, and different models for inferring hashing functions. However, there is little work integrating supervised and unsupervised hash codes learning into a single framework. Moreover, the hash function learning component is usually based on hand-crafted visual features extracted from the training images. The performance of a content-based image retrieval system crucially depends on the feature representation and such hand-crafted visual features may degrade the accuracy of the hash functions. In this paper, we propose a semi-supervised deep learning hashing (DLH) method for fast multimedia retrieval. More specifically, in the first component, we utilize both visual and label information to learn an relative similarity graph that can more precisely reflect the relationship among training data, and then generate the hash codes based on the graph. In the second stage, we apply a deep convolutional neural network (CNN) to simultaneously learn a good multimedia representation and hash functions. Extensive experiments on three popular datasets demonstrate the superiority of our DLH over both supervised and unsupervised hashing methods.

Robust video watermarking based on affine invariant regions in the compressed domain

This paper proposes a novel robust video watermarking scheme based on local affine invariant features in the compressed domain. This scheme is resilient to geometric distortions and quite suitable for DCT-encoded compressed video data because it performs directly in the block DCTs domain. In order to synchronize the watermark, we use local invariant feature points obtained through the Harris-Affine detector which is invariant to affine distortions. To decode the frames from DCT domain to the spatial domain as fast as possible, a fast inter-transformation between block DCTs and sub-block DCTs is employed and down-sampling frames in the spatial domain are obtained by replacing each sub-blocks DCT of 2x2 pixels with half of the corresponding DC coefficient. The above-mentioned strategy can significantly save computational cost in comparison with the conventional method which accomplishes the same task via inverse DCT (IDCT). The watermark detection is performed in spatial domain along with the decoded video playing. So it is not sensitive to the video format conversion. Experimental results demonstrate that the proposed scheme is transparent and robust to signal-processing attacks, geometric distortions including rotation, scaling, aspect ratio changes, linear geometric transforms, cropping and combinations of several attacks, frame dropping, and frame rate conversion.

A novel image copy detection scheme based on the local multi-resolution histogram descriptor

The conventional research on the image copy detection concentrates on extracting features which are robust enough to resist various kinds of image attacks. However, the global features are sensitive to geometric attacks, especially cropping and rotation, while the local features cannot substantially represent the image spatial information and structure context. Instead of simply extracting feature from local region or global image directly, we propose a novel image copy detection scheme based on Scale Invariant Feature Transform (SIFT) detector and multi-resolution histogram descriptor (MHD). In this novel algorithm, a series of robust, homogenous and large size circular patches are firstly constructed using the SIFT detector, and then the MHD is introduced to generate a discriminative feature vector for each patch. Experimental results obtained from the benchmark attacks demonstrate that the performance of the proposed approach is better than existing methods, especially on the test against geometric distortions.

Fast and robust video copy detection scheme using full DCT coefficients

In this paper, a fast and robust video copy detection scheme is proposed, which is suitable for the DCT-coded video sequences. To address the efficiency and effectiveness issue, we extract the video signature directly from the compressed domain. The video sequence clusters are constructed with a fixed length. Each cluster consists of several fictional key-frames. For each key-frame, some low-middle frequency full DCT coefficients are obtained directly from block DCT coefficients, and their ordinal measure is computed and acts as video signature. A rotation compensational strategy is further employed to resist the rotation attacks. The experimental results show that the proposed scheme can be resilient to various types of video transformations, including scaling, rotation, speed change, text insertion, and subsequence insertion/deletion etc.. The most important thing is that the proposed approach not only handles geometric distortion perfectly, but also reduces the computation costs substantially.

Adaptive Image Watermarking Algorithm in Contourlet Domain

In this paper, a novel watermarking algorithm using contourlet transform is proposed. The proposed algorithm takes advantage of a multiscale framework and directionality to extract the significant texture information of an image from its contourlet domain. By preserving both the robust property and the imperceptibility before and after watermarking, an adaptive watermarking scheme based on texture and luminance features in contourlet domain is proposed, which use the texture and luminance features of the image to find the embedding watermarking position. Finally, the analytical expressions are contrasted with experimental results where the robustness of the proposed watermarking system is evaluated against standard watermarking attacks.

Deep and fast: Deep learning hashing with semi-supervised graph construction*

Abstract Learning-based hashing methods are becoming the mainstream for approximate scalable multimedia retrieval. They consist of two main components: hash codes learning for training data and hash functions learning for new data points. Tremendous efforts have been devoted to designing novel methods for these two components, i.e., supervised and unsupervised methods for learning hash codes, and different models for inferring hashing functions. However, there is little work integrating supervised and unsupervised hash codes learning into a single framework. Moreover, the hash function learning component is usually based on hand-crafted visual features extracted from the training images. The performance of a content-based image retrieval system crucially depends on the feature representation and such hand-crafted visual features may degrade the accuracy of the hash functions. In this paper, we propose a semi-supervised deep learning hashing (DLH) method for fast multimedia retrieval. More specifically, in the first component, we utilize both visual and label information to learn an optimal similarity graph that can more precisely encode the relationship among training data, and then generate the hash codes based on the graph. In the second stage, we apply a deep convolutional network to simultaneously learn a good multimedia representation and a set of hash functions. Extensive experiments on five popular datasets demonstrate the superiority of our DLH over both supervised and unsupervised hashing methods.

A new fingerprinting scheme using social network analysis for majority attack

Digital fingerprinting is a technology to protect multimedia content from unauthorized redistribution. However, collusion attack is a cost-efficient attack for digital fingerprinting, where groups of dishonest users create a pirate copy using their copies for the purpose of attenuating or removing the fingerprints. In this paper, FCBSN, Fingerprinting Code Based on Social Networks, for coding the user’s fingerprints to resist majority attack, is proposed. The proposed scheme stems from the concept of coalition which always occurred in a social network. Different from all existing work, we explore the notion of the hierarchical community structure of social network and its intrinsic properties to assign fingerprints to users, drawing on the social relation according to the similar metric between two users. Theoretical analysis and experimental results show that the FCBSN detector outperforms the existing group detector for BS code and Tardos detector by large margins.

Robust localized image watermarking based on invariant regions

The robustness of the localized watermarking methods mainly depends on the robustness of the feature locating the watermark. Based on the mean luminance of the disk, a rotation and scale invariant feature extraction algorithm is proposed. A theoretical verification of the rotation and scale invariance of extracted feature points in the continuous image is further performed. The extracted feature points are used to construct rotation and scale invariant circular regions, where the watermark is embedded after affine normalization. Experimental results show that the constructed regions fit the watermarking applications much better than those in previous feature-based watermarking schemes from the aspect of robustness against common attacks including filtering, JPEG compression, cropping, rotation and scaling, and the proposed localized image watermarking scheme has better robustness than previous feature-based watermarking schemes against common signal process and geometrical attacks while maintaining imperceptibility.