Search

End-to-end learned image compression codecs have notably emerged in recent years. These codecs have demonstrated superiority over conventional methods, showcasing remarkable flexibility and adaptability across diverse data domains while supporting new distortion losses. Despite challenges such as computational complexity, learned image compression methods inherently align with learning-based data processing and analytic pipelines due to their well-suited internal representations. The concept of Video Coding for Machines has garnered significant attention from both academic researchers and industry practitioners. This concept reflects the growing need to integrate data compression with computer vision applications. In light of these developments, we present a comprehensive survey and review of lossy image compression methods. Additionally, we provide a concise overview of two prominent international standards, MPEG Video Coding for Machines and JPEG AI. These standards are designed to bridge the gap between data compression and computer vision, catering to practical industry use cases. [ABSTRACT FROM AUTHOR]

In this study, we present a federated learning approach that combines a multi-branch network and the Oort client selection algorithm to improve the performance of federated learning systems. This method successfully addresses the significant issue of non-iid data, a challenge not adequately tackled by the commonly used MFedAvg method. Additionally, one of the key innovations of this research is the introduction of uniformity, a metric that quantifies the disparity in training time amongst participants in a federated learning setup. This novel concept not only aids in identifying stragglers but also provides valuable insights into assessing the fairness and efficiency of the system. The experimental results underscore the merits of the integrated multi-branch network with the Oort client selection algorithm and highlight the crucial role of uniformity in designing and evaluating federated learning systems. [ABSTRACT FROM AUTHOR]

This work presents an efficient and effective system allowing hospitals to share patients' private information while ensuring that each hospital database's medical records will not be leaked; moreover, the privacy of patients who access the data will also be protected. We assume that the thread model of the hospital's security is semi-honest (i.e., curious but honest), and each hospital hired a trusted medical records department administrator to manage patients' private information from other hospitals. With the help of Homomorphic Encryption- and Private Set Intersection -related algorithms, our proposed system protects patient privacy, allows physicians to obtain patient information across hospitals, and prevents threats such as troublesome insider attacks and man-in-the-middle attacks. [ABSTRACT FROM AUTHOR]

Recently, federated learning (FL) has gradually become an important research topic in machine learning and information theory. FL emphasizes that clients jointly engage in solving learning tasks. In addition to data security issues, fundamental challenges in this type of learning include the imbalance and non-IID among clients' data and the unreliable connections between devices due to limited communication bandwidths. The above issues are intractable to FL. This study starts from the uncertainty analysis of deep neural networks (DNNs) to evaluate the effectiveness of FL, and proposes a new architecture for model aggregation. Our scheme improves FL's performance by applying knowledge distillation and the DNN's uncertainty quantification methods. A series of experiments on the image classification task confirms that our proposed model aggregation scheme can effectively solve the problem of non-IID data, especially when affordable transmission costs are limited. [ABSTRACT FROM AUTHOR]

Secure comparison is a fundamental problem in multiparty computation. There are two different parties, each holding an  l -bit integer, denoted by  a  and  b , respectively. The goal of secure comparison is to compute the order relationship between  a  and  b , say  (a > b) ∈ { 0 , 1 } , without revealing their inputs to any others. Since previous solutions based on homomorphic encryption need at least  Ω (l)   encryptions for each  l -bit comparison, the total encryption time leads to a computational bottleneck for these protocols. This work presents a fast, semi-honest, secure comparison protocol based on the BFV encryption scheme. With its vector-like plaintext space, the number of required encryptions can be significantly reduced; actually, only six encryptions are needed for each comparison in our protocol. In other words, the proposed protocol can achieve the time complexity  O ˜ (λ + l ) for a given security parameter λ. As a result, 4096-bit integers can be securely compared within 12.08 ms, which is 280 times faster than the state-of-the-art homomorphic encryption-based secure comparison protocol. Furthermore, we can compare k pairs of  l ⋅ k − 1 -bit integers with almost the same execution time as comparing  l -bit integers and achieve higher throughput regardless of the compared integer size. [ABSTRACT FROM AUTHOR]

Steganography is one of the most crucial methods for information hiding, which embeds secret data on an ordinary file or a cover message for avoiding detection. We designed a novel rate-distortion-based large-capacity secure steganographic system, called rate-distortion-based Stego (RD-Stego), to effectively solve the above requirement. The considered effectiveness of our system design includes embedding capacity, adaptability to chosen cover attacks, and the stability of the trained model. The proposed stego scheme can hide multiple three-channel color images and QR codes within another three-channel color image with low visual distortion. Empirically, with a certain degree of robustness against the chosen cover attack, we state that the system offers up to 192+ bits-per-pixel (bpp) embedding of a payload and leaks no secret-related information. Moreover, to provide theoretical foundations for our cost function design, a mutual information-based explanation of the choices of regulation processes is herein included. Finally, we justify our system's claimed advantages through a series of experiments with publicly available benchmark datasets. [ABSTRACT FROM AUTHOR]

The Asymmetric Numeral System (ANS) is a new entropy compression method that the industry has highly valued in recent years. ANS is valued by the industry precisely because it captures the benefits of both Huffman Coding and Arithmetic Coding. Surprisingly, compared with Huffman and Arithmetic coding, systematic descriptions of ANS are relatively rare. In 2017, JPEG proposed a new image compression standard—JPEG XL, which uses ANS as its entropy compression method. This fact implies that the ANS technique is mature and will play a kernel role in compressing digital images. However, because the realization of ANS involves combination optimization and the process is not unique, only a few members in the compression academia community and the domestic industry have noticed the progress of this powerful entropy compression approach. Therefore, we think a thorough overview of ANS is beneficial, and this idea brings our contributions to the first part of this work. In addition to providing compact representations, ANS has the following prominent feature: just like its Arithmetic Coding counterpart, ANS has Chaos characteristics. The chaotic behavior of ANS is reflected in two aspects. The first one is that the corresponding compressed output will change a lot if there is a tiny change in the original input; moreover, the reverse is also applied. The second is that ANS compressing an image will produce two intertwined outcomes: a positive integer (aka. state) and a bitstream segment. Correct ANS decompression is possible only when both can be precisely obtained. Combining these two characteristics helps process digital images, e.g., art collection images and medical images, to achieve compression and encryption simultaneously. In the second part of this work, we explore the characteristics of ANS in depth and develop its applications specific to joint compression and encryption of digital images. [ABSTRACT FROM AUTHOR]

Currently, cloud computing has become increasingly popular and thus, many people and institutions choose to put their data into the cloud instead of local environments. Given the massive amount of data and the fidelity of cloud servers, adequate security protection and efficient retrieval mechanisms for stored data have become critical problems. Attribute-based encryption brings the ability of fine-grained access control and can achieve a direct encrypted data search while being combined with searchable encryption algorithms. However, most existing schemes only support single-keyword or provide no ranking searching results, which could be inflexible and inefficient in satisfying the real world's actual needs. We propose a flexible multi-keyword ranked searchable attribute-based scheme using search trees to overcome the above-mentioned problems, allowing users to combine their fuzzy searching keywords with AND–OR logic gates. Moreover, our enhanced scheme not only improves its privacy protection but also goes a step further to apply a semantic search to boost the flexibility and the searching experience of users. With the proposed index-table method and the tree-based searching algorithm, we proved the efficiency and security of our schemes through a series of analyses and experiments. [ABSTRACT FROM AUTHOR]

Privacy protection data processing has been critical in recent years when pervasively equipped mobile devices could easily capture high-resolution personal images and videos that may disclose personal information. We propose a new controllable and reversible privacy protection system to address the concern in this work. The proposed scheme can automatically and stably anonymize and de-anonymize face images with one neural network and provide strong security protection with multi-factor identification solutions. Furthermore, users can include other attributes as identification factors, such as passwords and specific facial attributes. Our solution lies in a modified conditional-GAN-based training framework, the Multi-factor Modifier (MfM), to simultaneously accomplish the function of multi-factor facial anonymization and de-anonymization. It can successfully anonymize face images while generating realistic faces satisfying the conditions specified by the multi-factor features, such as gender, hair colors, and facial appearance. Furthermore, MfM can also de-anonymize de-identified faces to their corresponding original ones. One crucial part of our work is design of physically meaningful information-theory-based loss functions, which include mutual information between authentic and de-identification images and mutual information between original and re-identification images. Moreover, extensive experiments and analyses show that, with the correct multi-factor feature information, the MfM can effectively achieve nearly perfect reconstruction and generate high-fidelity and diverse anonymized faces to defend attacks from hackers better than other methods with compatible functionalities. Finally, we justify the advantages of this work through perceptual quality comparison experiments. Our experiments show that the resulting LPIPS (with a value of 0.35), FID (with a value of 28), and SSIM (with a value of 0.95) of MfM demonstrate significantly better de-identification effects than state-of-the-art works. Additionally, the MfM we designed can achieve re-identification, which improves real-world practicability. [ABSTRACT FROM AUTHOR]

With the advance of 3D display and related technologies, 3D video applications have been booming up recently. The color plus depth-map 3D video gains the attention for its relatively low storage and transmission bandwidth requirements as compared with stereo and multi-view videos. The color plus depth-map 3D video provides the depth information about the recorded content which brings new rooms for video processing and compression. This work reveals the usage of depth information for speeding up video encoding process. A novel fast mode decision algorithm based on depth information is proposed. The algorithm is also applicable to videos in which the associated depth-maps are not the ground-truth. By utilizing the depth information, we can classify scenes according to the associated depth and assign the appropriate modes. We can also deduce the configuration of the given macroblock by the associated depth information and select the suitable mode candidates; therefore, the encoding process will be speeded-up accordingly. Experimental results show that the proposed algorithm is faster than existing mode decision approaches with similar coding efficiency. The entire encoding time is reduced by about 63% for Standard Definition (SD) and High Definition (HD) formats while the Rate-Distortion (RD) performance is almost the same as compared with the exhaustive full mode decision approach. [ABSTRACT FROM AUTHOR]

The content protection for image-based 3D data is getting more importance with the advance of low cost 3D display devices. The depth-image-based rendering (DIBR) 3D image is one of the image-based 3D data which consists of the center image and the depth image generated by the content provider. The left-eye image and the right-eye image are rendered from the center image and the depth image at the content consumer side. The blind watermarking for DIBR 3D image is rarely studied in the literature. In this paper, a novel blind multiple watermarking scheme is proposed to deal with the content protection problem of DIBR 3D images. Besides the usual requirement of mutual orthogonality among reference patterns for multiple watermark embedding, we found that proper embedding order plays an even more important role in watermarking the DIBR 3D images. Experimental results show that the proposed scheme is robust against the JPEG compression and noise adding attacks. More interestingly, it is found that the proposed watermarking can also tolerate large range variations of the depth image during rendering. [ABSTRACT FROM AUTHOR]

Abstract: This paper presents a feasible system for multimedia fingerprinting. One of the important problems of fingerprinting is watermarking strategies for the fingerprints, in other words, where to embed and how to embed. We address these two major problems of fingerprint-watermarking for multimedia in the proposed content-adaptive watermarking scheme. First, the fingerprint codes are often very long if the applicable resisting ability to colluders and large customer bases are needed. As a result, the acute degradation of fingerprinted content may be unacceptable. We design a strength-decision algorithm, on the basis of neural networks, to adaptively embed the long-length fingerprints with suitable magnitudes to different image regions. This adaptive watermarking technique maintains the equilibrium of the robustness and the imperceptibility without the effort to deal with visual models. Second, we analyze the disturbance of collusion attacks on images and propose an optimization algorithm which can select better embedding positions to resist collusion attacks and preserve acceptable transparency of the watermark according to different multimedia contents. In addition, we consider the lossy property of multimedia watermarking and use a sequential detection strategy to identify colluders, which can tolerate erasures and errors possibly induced in the watermarking process or communication channel. Experimental results show the high detection correctness of traitor tracing. It implies that our fingerprinting system, constructed by applying c-TA code to the content-adaptive watermarking scheme and a sequential detection algorithm, is effective for multimedia application. One can replace the fingerprint codes in our system with other existing codes to obtain effective fingerprinting systems with higher tracing correctness and practical parameters (reasonable collusion-resilience and applicable size of customer bases). [Copyright &y& Elsevier]

Genetic algorithms have been applied to many optimization and search problems and shown to be very efficient. However, the efficiency of genetic algorithms is not guaranteed in those applications where the search space is small, such as the block motion estimation in video coding applications, or equivalently the chromosome length is relatively short, less than 5 for example. Since the characteristics of these small search space applications are far away from the conventional search problems in which the common genetic algorithms worked well. New treatments of genetic algorithms for dealing with the small range search problems are therefore of interest. In this paper, the efficiency (will be defined later) of the genetic operations of common genetic algorithms, such as crossover and mutation, are analyzed for this special situation. As expected, the so-obtained efficiency/performance of the genetic operations is quite different from that of their traditional counterparts. To fill this gap, a lightweight genetic search algorithm is presented to provide an efficient way for generating near optimal solutions for these kinds of applications. The control overheads of the lightweight genetic search algorithm is very low as compared with that of the conventional genetic algorithms. It is shown by simulations that a lot of computations can be saved by applying the newly proposed algorithm while the search results are still well approved. [ABSTRACT FROM AUTHOR]

This paper presents a class of neural networks suitable for the application of decoding error-correcting codes. The neural model is basically a perceptron with a high-order polynomial as its discriminant function. A single layer of high-order perceptrons is shown to be able to decode a binary linear block code with at most 2[SUPm] weights in each perceptron, where m is the parity length. For some subclass codes, the number of weights needed can be much less. The (2[SUPm]-l,2[SUPm]-1-m) Hamming code can be decoded with only m+l weights in each perceptron. With the help of genetic algorithms, efficient neural decoders with 2t+l terms for each bit for some t-error correctable cyclic and BCH codes are obtained. The neural decoders are formulated as a set of parity networks in the first layer followed by a linear perceptron in the second layer, and thus have simple implementations in analog VLSI technology. [ABSTRACT FROM AUTHOR]

Presents information on a study which proposed an efficient algorithm to construct a symmetrical reversible variable length code (RVLC) from a given Huffman code. Backgrounds of symmetrical RVLC; Algorithm with an efficient codeword selection mechanism; Bounds of symmetrical RVLC; Experimental results.

Proposes a hybrid approach to eliminate the drift in an H.263-based video codec with two-layer scalability. Review of the discrete cosine transform (DCT)-domain layered video coding techniques; Why the drift occurred; Design and implementation issues of the proposed codec; Experimental results.

Presents information on a study that used vertically invariant morphological (MV) filters for time-varying or adaptive signal processing. Description of MV openings and closings in adaptive signal filtering; How the progressive umbra-filling procedure (PUF) was applied to the area of adaptive signal filtering; Conclusions.

Presents an index permutation-based fast two-dimensional discrete cosine transform (2-D DCT) algorithm. Illustration of the refined fast algorithm for computing the 2-D DCT; Complexity analysis of the post-addition stages; Conclusions.

Looks at the development of a multiresolution-based watermarking technique used to enclose digital watermarks into images, in image processing. In-depth look at the technique; Effectiveness of the technique; Decomposition of the multiresolution signal; Discussion on data distribution.

Presents a study which proposed an automatic facial feature recognition algorithm. Face region estimation and feature extraction stages of the algorithm; Simulation results; Conclusion.

Presents information on a study which focused on a proposed image authentication technique by embedding digital watermarks into images. Description of the embedding methods; Information on the watermark extraction method; Experimental results; Issues regarding the proposed method; Conclusions.

Presents information on a study which proposed the use of a lightweight genetic search algorithm (LGSA) for video coding, to meet the need of block-matching processing. Comparison of the LGSA and full search algorithm (FSA); Details on the use of genetic algorithms; Findings of the study.

Describes a fast computation algorithm for the two-dimensional discrete cosine transform (2-D DCT). Capabilities of the algorithm; Complexity analysis of the post-addition stage.

Presents information on the proposed MPEG-based technique for embedding digital `watermarks' into digital video. Method used to hide secret information into the signals to discourage unauthorized copying or attest the origin of the media; Advantages of the proposed method; Discussion on the experimental results.

Presents a real time personal computer-based software implementation of H.261 video codec. How to develop a real time H.261 decoder; Most important issues in video encoding; Performance of the proposed software H.261 video codec; Conclusion.

Proposes a content-based bit-rate control scheme to find the main attributes of each macroblock in video coding. Overview of the system structure; Extraction of active and face regions of the scheme; Genetic search of facial features; Rate control scheme to meet the designated bit-rate constraint; Experimental results; Conclusion.

Proposes a block matching algorithm with a dynamic search range algorithm (DSRA) to improve the efficiency of an encoding system and make the software-based real-time video encoding more feasible. Derivation of the DSRA; Experimental results and discussion of the study; Conclusion.

Proposes the application of wavelet transform as a tool for image mosaic. Application of wavelet transform; Combination of signals; Information content analysis of signals; Emergence of wavelet theory.

Presents the Popular Video Coder II (PVC II), a generation of real-time software-based video compression equipment. Comparison of PVC II and traditional video coder; Improved coding performance; Absence of compression hardware; Encoding and decoding of video data; Multimedia system; Advantages of PVC II.

Describes a low-cost and low-bit-rate editing system which produces Motion Picture Experts Group-1 (MPEG-1) decodable bistreams. System considerations; Skeleton of the transform domain JPEG to MPEG-1 format converter; Bit-rate reduction; Performance of the proposed algorithms.

In theory, high key and high plaintext sensitivities are a must for a cryptosystem to resist the chosen/known plaintext and the differential attacks. High plaintext sensitivity can be achieved by ensuring that each encrypted result is plaintext-dependent. In this work, we make detailed cryptanalysis on a published chaotic map-based image encryption system, where the encryption process is plaintext Image dependent. We show that some designing flaws make the published cryptosystem vulnerable to chosen-plaintext attack, and we then proposed an enhanced algorithm to overcome those flaws. [ABSTRACT FROM AUTHOR]

Proposes a generalized output pruning algorithm for matrix-vector multiplication. Operation of a general matrix-vector multiplication; Details on the encoding and decoding procedures in the proposed pruning algorithm; Application of the proposed output algorithm to the computation of pruning discrete cosine transform.

Shows that the discrete cosine transform (DCT) can be obtained by projecting the discrete Fourier transform from the extension field to the basefield. Interpretation of the discrete cosine transform; Recursive fast DCT algorithm; Conclusions.

Highlights: [•] The depth information of 2D paintings can only be generated by users. [•] Inspired by sculpturing, we address the problem as an iterative process. [•] We expect this work can inspire more applications of 2D painting plus depth. [Copyright &y& Elsevier]

Abstract: Pipelining has been applied in many area to improve system performance by overlapping executions of hardware or software computing stages. However, direct pipelining for H.264 decoding is difficult because video bitstreams are encoded with lots of dependencies and little parallelism is left to be explored. Fortunately, pure software pipelining can still be applied to H.264 decoding at macroblock level with reasonable performance gain. However, the pipeline stages might need to synchronize with each other and incur lots of extra overhead. For optimized decoders, the overhead is relatively more significant and software pipelining might lead to negative performance gain. We first group multiple stages into larger batches and execute these batches concurrently, called batch-pipelining, to explore more parallelism on multicore systems. Experimental results show that it can speed the decoding up to 89% and achieve up to 259 and 69 frames per second for resolution 720P and 1080P, respectively, on a 4-core ×86 machine over an optimized H.264 decoder. Because of its flexibility, batch-pipelining can be applied to not only H.264 but also many similar applications, such as the next-generation video coding: high efficiency video coding. Therefore, we believe the batch-pipelining mechanism creates a new effective direction for software codec development. [Copyright &y& Elsevier]

Abstract: Distributed video coding (DVC) is a new coding paradigm targeting on applications with the need of low-complexity and/or low-power encoding at the cost of a high-complexity decoding. In the DVC architectures based on Error Control Codes (ECCs) with a feedback channel, the high decoding complexity comes from the decode-check-request iterations between the ECC encoder and the ECC decoder. In this paper, a parallel message-passing decoding algorithm for computing low density parity check (LDPC) syndromes is applied through the Compute Unified Device Architecture (CUDA) based on General Purpose Graphics Processing Unit (GPGPU). Furthermore, we proposed a novel rate control mechanism, dubbed as the Ladder Step Size Request (LSSR), to reduce the number of requests which leads to much speedup gain. Experimental results show that, through our work, the overall DVC decoding speedup gain can reach 46.52 with only 0.2dB rate distortion performance loss. [Copyright &y& Elsevier]

Describes the Perceptual Audio Coder (PAC), an audio coding technology developed by Dolby Company. Its features; Capabilities; Approaches to accelerate the decoder.

Presents the result of a study on a prototype of MPEG-4-like video encoder and a prototype of Java-based video encoder which can be executed on the Internet. Background information on MPEG-4 Video; Details on the MPEG-4 like video encoder and the MPEG-4 like video decoder.

In this work, three techniques for enhancing various chaos-based joint compression and encryption (JCAE) schemes are proposed. They respectively improved the execution time, compression ratio, and estimation accuracy of three different chaos-based JCAE schemes. The first uses auxiliary data structures to significantly accelerate an existing chaos-based JCAE scheme. The second solves the problem of huge multidimensional lookup table overheads by sieving out a small number of important sub-tables. The third increases the accuracy of frequency distribution estimations, used for compressing streaming data, by weighting symbols in the plaintext stream according to their positions in the stream. Finally, two modified JCAE schemes leveraging the above three techniques are obtained, one applicable to static files and the other working for streaming data. Experimental results show that the proposed schemes do run faster and generate smaller files than existing JCAE schemes, which verified the effectiveness of the three newly proposed techniques. [ABSTRACT FROM AUTHOR]

Detecting humans in crowded environment is profitable but challenging in video surveillance. We propose an efficient human detection method by combining both motion and appearance clues. Moving pixels are first extracted by background subtraction, and then a filtering step is used to narrow the range for human template matching. We utilize integral images to fast generate shape information from edge maps of each frame and define the matching probability to be capable of detecting both full-body and partial-body. Representative human templates are constructed by sparse contours on the basis of the point distribution model. Moreover, linear regression analysis is also applied to adaptively adjust the template sizes. With the aid of the proposed foreground ratio filtering and the multi-sized template matching techniques, experimental results show that our method not only can efficiently detect humans in a crowded environment, but also largely enhance the resultant detection accuracy. [ABSTRACT FROM AUTHOR]

Abstract: How to effectively and less-intrusively deliver the advertising message by spatially replacing regions with advertisements in a period of exposure time for sports videos has been known as a challenging problem. The size, placement locations, and the representation of advertisement are the critical factors that have significant impact on both the recognition effectiveness and the perceived intrusiveness. In this paper, we take advertising theory, psychology, and computational aesthetics into account to develop a novel virtual advertising mechanism, called virtual spotlighted advertising (ViSA), for tennis videos. We utilize the extraneous visual acuity of viewers while watching the attractive object, such that they are not much disturbed from the progress of the game, and at the same time, the inserted advertisement can effectively deliver its message across to them. We propose a framework and realize an exemplary system to serve ViSA. The system automatically detects the candidate insertion points in both temporal and spatial domains and estimates the most effective region for visual communication. Then, the harmonically re-colored advertisements with foveation model based non-uniform transparency, are projected onto the court. The evaluation results demonstrate the effectiveness of the proposed ViSA in terms of recall and recognition. Moreover, the induced visual intrusiveness is limited by the proposed innovative representation style. [Copyright &y& Elsevier]

There are four major components in a generic digital home framework: media center, set-top box, media adapter and portable viewer. In this paper, we present a DSP- based multi-format video decoding engine for realizing media adapter application. By choosing a powerful DSP core and designing an efficient and configurable software engine, the proposed solution is believed to possess low-cost and high- flexibility advantages at the same time. In our current implementation, the video decoding engine can decode most of the major video formats in the market, such as MPEG- I/2/4, H.264/AVC baseline/Main, VC-1 and its precedents, and some other MPEG-4-like formats. Simulation results demonstrate the efficiency of the decoding engine: videos up to SD resolution can be decoded seamlessly. The engine has been successfully integrated into an available media adapter module and the corresponding product. Our future work is to enhance the engine performance capable of handling HD contents. [ABSTRACT FROM AUTHOR]