Bøger udgivet af now publishers Inc

In this monograph, an overview of recent developments and the state-of-the-art in image/video restoration and super-resolution (SR) using deep learning is presented. Deep learning has made a significant impact, not only on computer vision and natural language processing but also on classical signal processing problems such as image/video restoration/SR and compression. Recent advances in neural architectures led to significant improvements in the performance of learned image/video restoration and SR. An important benefit of data-driven deep learning approaches is that neural models can be optimized for any differentiable loss function, including visual perceptual loss functions, leading to perceptual video restoration and SR, which cannot be easily handled by traditional model-based approaches. The publication starts with a problem statement and a short discussion on traditional vs. data-driven solutions. Thereafter, recent advances in neural architectures are considered, and the loss functions and evaluation criteria for image/video restoration and SR are discussed. Also considered are the learned image restoration and SR, as learning either a mapping from the space of degraded images to ideal images based on the universal approximation theorem, or a generative model that captures the probability distribution of ideal images. Practical problems in applying supervised training to real-life restoration and SR are also included, as well as the solution models. In the section on learned video SR, approaches to exploit temporal correlations in learned video processing are covered, and then the perceptual optimization of the network parameters to obtain natural texture and motion is discussed. A comparative discussion of various approaches concludes the publication.

Methods for image recovery and reconstruction aim to estimate a good-quality image from noisy, incomplete, or indirect measurements. Such methods are also known as computational imaging. New methods for image reconstruction attempt to lower complexity, decrease data requirements, or improve image quality for a given input data quality. Image reconstruction typically involves optimizing a cost function to recover a vector of unknown variables that agrees with collected measurements and prior assumptions. State-of-the-art image reconstruction methods learn these prior assumptions from training data using various machine learning techniques, such as bilevel methods. This review discusses methods for learning parameters for image reconstruction problems using bilevel formulations, and it lies at the intersection of a specific machine learning method, bilevel, and a specific application, filter learning for image reconstruction. The review discusses multiple perspectives to motivate the use of bilevel methods and to make them more easily accessible to different audiences. Various ways to optimize the bilevel problem are covered, providing pros and cons of the variety of proposed approaches. Finally, an overview of bilevel applications in image reconstruction is provided.

Rank-metric codes date back to the 1970s and today play a vital role in many areas of coding theory and cryptography. In this survey the authors provide a comprehensive overview of the known properties of rank-metric codes and their applications. The authors begin with an accessible and complete introduction to rank-metric codes, their properties and their decoding. They then discuss at length rank-metric code-based quantum resistant encryption and authentication schemes. The application of rank-metric codes to distributed data storage is also outlined. Finally, the constructions of network codes based on MRD codes, constructions of subspace codes by lifting rank-metric codes, bounds on the cardinality, and the list decoding capability of subspace codes is covered in depth. Rank-Metric Codes and Their Applications provides the reader with a concise, yet complete, general introduction to rank-metric codes, explains their most important applications, and highlights their relevance to these areas of research.

Image Alignment and Stitching: A Tutorial reviews image alignment and image stitching algorithms. Image alignment algorithms can discover the correspondence relationships among images with varying degrees of overlap. They are ideally suited for applications such as video stabilization, summarization, and the creation of panoramic mosaics. Image stitching algorithms take the alignment estimates produced by such registration algorithms and blend the images in a seamless manner, taking care to deal with potential problems such as blurring or ghosting caused by parallax and scene movement as well as varying image exposures. Image Alignment and Stitching: A Tutorial reviews the basic motion models underlying alignment and stitching algorithms, describes effective direct (pixel-based) and feature-based alignment algorithms, and describes blending algorithms used to produce seamless mosaics. It closes with a discussion of open research problems in the area. Image Alignment and Stitching: A Tutorial is an invaluable resource for anyone planning or conducting research in this particular area, or computer vision generally. The essentials of the topic are presented in a tutorial style and an extensive bibliography guides towards further reading.

Offers new perspectives on advanced (cyber) security innovation (eco) systems covering key different perspectives. The book provides insights on new security technologies and methods for advanced cyber threat intelligence, detection and mitigation.

Reviews the information relaxation approach which works by reducing a complex stochastic Dynamic Programming to a series of scenario-specific deterministic optimization problems solved within a Monte Carlo simulation.

Lists the seminal and pioneering research efforts conducted by a limited group of scholars from different disciplines that challenged traditional thought on small business and entrepreneurship - these pioneers and their specific contributions transformed our thinking about entrepreneurs.

This book provides students, researchers and practitioners with a deep understanding of the theory of online auctions and gives practical examples of how to implement in modern-day internet systems.

Privacy and Consumer Empowerment in Online Advertising provides an overview of the different issues that are in play in consumer privacy and in empowering consumers with rights to manage the privacy of their data. The authors review the existing knowledge on this topic and discuss implications for consumers, for advertisers, and for ad serving platforms that enable advertisers to reach consumers.The introductory section provides an outline and briefly reviews the key ideas. Section 2 discusses the key aspects of the GDPR, the CCPA and the CPRA. Since the implementation of the GDPR in May 2018, some early empirical evidence has emerged of its impact and this is examined in Section 3. The authors review the privacy and economic frameworks in Section 4. Section 5 discusses the theoretical work in this area enhances our understanding of the impact of privacy regulation on consumers and on online advertising. Section 6 examines how consumers are presented with privacy notices and their (in)ability to make privacy choices due to a variety of factors. Section 7 reviews how firms attach value to consumers' data. In light of the passing of privacy regulation, firms have been attempting to develop methods for privacy-preserving targeted advertising. In Section 8, we discuss some of these attempts such as FLoC and TURTLEDOVE, which aim to target consumers based on their interests and/or their website visit history, but without compromising their privacy. Finally, Section 9 concludes with a discussion.An overall summary is that privacy concerns have been heightened in the past two decades and this has led to the passing of privacy regulations addressing data security and privacy rights. After these regulations, a significant minority of consumers have chosen to not provide consent for their data to be collected, used and shared. However, most consumers still do not properly understand the key implications of privacy policies of firms, and more efforts are needed in that regard. Also, technologies are being developed for privacy-preserving user targeting. Finally, regarding firms, data frictions caused by privacy regulations have, in turn, caused negative consequences for small advertisers, publishers and service providers. The authors provide some directions for future work that may be valuable to move thinking forward on this increasingly important topic.

The comprehensive review of neurosymbolic programming introduces the reader to the topic and provides an insightful treatise on an increasingly important topic at the intersection of programming languages and machine learning.

This book is an introduction to Acceleration Methods used in convex optimization that enables the reader to quickly understand the important principles and apply the techniques to their own research.

This monograph reviews the work that HCI and other disciplines has produced on the digital transcendent experience and combines it with an illustrated design fiction.

Presents some existing and new results on analysis and design of finite-time and fixed-time converging systems. Two main groups of approaches for analysis/synthesis of this kind of convergence, Lyapunov functions and the theory of homogeneous systems, are considered.

Reviews research on the design and evaluation of search user interfaces of the past 10 years. It integrates state-of-the-art research in the areas of information seeking behavior, information retrieval, and human-computer interaction on the topic of search interface.

This monograph focusses on the current research activities and emerging trends that relate to numerous technological functionalities, and it should be of interest to students, researchers and engineers active in the fields related to Circuits and Systems for Biomedical Engineering.

Variational autoencoders (VAEs) are powerful deep generative models widely used to represent high-dimensional complex data through a low-dimensional latent space learned in an unsupervised manner. In this volume, the authors introduce and discuss a general class of models, called dynamical variational autoencoders.

The book is a self-contained treatment of a topic that sits at the intersection of machine learning and formal verification. It can serve as an introduction to the field for first-year graduate students or senior undergraduates, even if they have not been exposed to deep learning or verification.