Current Trends on the Acquisition, Virtual Representation, and Interaction of Cultural Heritage

Cultural heritage (CH) tells us about our roots, and therefore, constitutes a rich value for the society. Its conservation, dissemination, and understanding are of utmost importance. In order to preserve CH for the upcoming generations, it needs to be documented, a process that nowadays is done digitally. Current trends involve a set of technologies (cameras, scanners, etc.) for the shape and radiometric acquisition of assets. Also, intangible CH can be digitally documented in a variety of forms. Having such assets virtualized, a proper dissemination channel is of relevance, and recently, new technologies that make use of interaction paradigms have emerged. Among them, in this chapter, the authors focus their attention in the technologies of virtual reality (VR), augmented reality (AR), and serious games (SGs). They aim to explore these technologies in order to show their benefits in the dissemination and understanding of CH. Though the work involving them is not trivial, and usually a multidisciplinary team is required, the benefits for CH make them worth it.

Recent Advances in Web3D Semantic Modeling

The implementation of virtual and augmented reality environments on the web requires integration between 3D technologies and web technologies, which are increasingly focused on collaboration, annotation, and semantics. Thus, combining VR and AR with the semantics arises as a significant trend in the development of the web. The use of the Semantic Web may improve creation, representation, indexing, searching, and processing of 3D web content by linking the content with formal and expressive descriptions of its meaning. Although several semantic approaches have been developed for 3D content, they are not explicitly linked to the available well-established 3D technologies, cover a limited set of 3D components and properties, and do not combine domain-specific and 3D-specific semantics. In this chapter, the authors present the background, concepts, and development of the Semantic Web3D approach. It enables ontology-based representation of 3D content and introduces a novel framework to provide 3D structures in an RDF semantic-friendly format.

Single View 3D Face Reconstruction

3D face reconstruction from a single 2D image is a fundamental computer vision problem of extraordinary difficulty that dates back to the 1980s. Briefly, it is the task of recovering the three-dimensional geometry of a human face from a single RGB image. While the problem of automatically estimating the 3D structure of a generic scene from RGB images can be regarded as a general task, the particular morphology and non-rigid nature of human faces make it a challenging problem for which dedicated approaches are still currently studied. This chapter aims at providing an overview of the problem, its evolutions, the current state of the art, and future trends.

An Overview on 3D Site Modelling in Civil Engineering

In past decades, for developing a site, engineers used the process of creating a scale model in order to determine their behaviour and to sketch the details collected manually using the drafting process, which behaves as a referring material during the construction of structures. Due to the boom in technology and limitations in drafting, the drawings have been digitized using computer-aided design (CAD) software as a two-dimensional structure (2D). Currently, these drawings are detailed as a three-dimensional structure (3D) that is briefly noted as 3D modelling. Three-dimensional site modelling is an active area that is involved in research and development of models in several fields that has been originated from the scale modelling. In this chapter, the topic 3D site modelling in civil engineering is discussed. First of all, the basic concepts of scale modelling, architectural modelling, and structural modelling are discussed. Then the concept of virtual-based 3D site modelling, its importance, benefits, and steps involved in site modelling are briefed.

Tropical Tree Species 3D Modelling and Classification Based on LiDAR Technology

Over the last decade or so, laser scanning technology has become an increasingly popular and important tool for forestry inventory, enabling accurate capture of 3D information in a fast and environmentally friendly manner. To this end, the authors propose here a system for tropical tree species classification based on 3D scans of LiDAR sensing technology. In order to exploit the interrelated patterns of trees, skeleton representations of tree point clouds are extracted, and their structures are divided into overlapping equal-sized 3D segments. Subsequently, they represent them as third-order sparse structure tensors setting the value of skeleton coordinates equal to one. Based on the higher-order tensor decomposition of each sparse segment, they 1) estimate the mode-n singular values extracting intra-correlations of tree branches and 2) model tropical trees as linear dynamical systems extracting appearance information and dynamics. The proposed methodology was evaluated in tropical tree species and specifically in a dataset consisting of 26 point clouds of common Caatinga dry-forest trees.

Review on Various Machine Learning and Deep Learning Techniques for Prediction and Classification of Quotidian Datasets

Machine learning can be defined as the ability of a computer to learn and solve a problem without being explicitly coded. The efficiency of the program increases with experience through the task specified. In traditional programming, the program and the input are specified to get the output, but in the case of machine learning, the targets and predictors are provided to the algorithm make the process trained. This chapter focuses on various machine learning techniques and their performance with commonly used datasets. A supervised learning algorithm consists of a target variable that is to be predicted from a given set of predictors. Using these established targets is a function that plots targets to a given set of predictors. The training process allows the system to train the unknown data and continues until the model achieves a desired level of accuracy on the training data. The supervised methods can be usually categorized as classification and regression. This chapter discourses some of the popular supervised machine learning algorithms and their performances using quotidian datasets. This chapter also discusses some of the non-linear regression techniques and some insights on deep learning with respect to object recognition.

Emerging Trends in 3D Image Reconstruction and Modeling

In recent years, there is a demand for 3D content for computer graphics, communications, and virtual reality. 3D modelling is an emerging topic that is applied in so many real-world applications. The images are taken through camera at multiple angles and medical imaging techniques like CT scan and MRI are also used. From a set of images, intersection of these projection rays is considered to be the position for 3D point. This chapter discusses the construction of 3D images from multiple objects. Various approaches used for construction, triangulation method, challenges in building this model, and the application of 3D models are explained in this chapter.

Visualization and Analysis of 3D Images Using Data Mining Approaches

With the recent advancements in supercomputer technologies, large-scale, high-precision, and realistic model 3D simulations have been dominant in the field of solar-terrestrial physics, virtual reality, and health. Since 3D numeric data generated through simulation contain more valuable information than available in the past, innovative techniques for efficiently extracting such useful information are being required. One such technique is visualization—the process of turning phenomena, events, or relations not directly visible to the human eye into a visible form. Visualizing numeric data generated by observation equipment, simulations, and other means is an effective way of gaining intuitive insight into an overall picture of the data of interest. Meanwhile, data mining is known as the art of extracting valuable information from a large amount of data relative to finance, marketing, the internet, and natural sciences, and enhancing that information to knowledge.

Medical 3D Graphics With eXtensible 3D

As the healthcare enterprise is adopting novel imaging and health-assessment technologies, we are facing unprecedented requirements in information sharing, patient empowerment, and care coordination within the system. Medical experts not only within US, but around the world should be empowered through collaboration capabilities on 3D data to enable solutions for complex medical problems that will save lives. The fast-growing number of 3D medical ‘images' and their derivative information must be shared across the healthcare enterprise among stakeholders with vastly different perspectives and different needs. The demand for 3D data visualization is driving the need for increased accessibility and sharing of 3D medical image presentations, including their annotations and their animations. As patients have to make decisions about their health, empowering them with the right tools to understand a medical procedure is essential both in the decision-making process and for knowledge sharing.

Data Augmentation Using GANs for 3D Applications

Modern deep learning techniques have proven that they have the capacity to be successful in a wide area of domains and tasks, including applications related to 3D and 2D images. However, their quality depends on the quality and quantity of the data with which models are trained. As the capacity of deep learning models increases, data availability becomes the most significant. To counter this issue, various techniques are utilized, including data augmentation, which refers to the practice of expanding the original dataset with artificially created samples. One approach that has been found is the generative adversarial networks (GANs), which, unlike other domain-agnostic transformation-based methods, can produce diverse samples that belong to a given data distribution. Taking advantage of this property, a multitude of GAN architectures has been leveraged for data augmentation applications. The subject of this chapter is to review and organize implementations of this approach on 3D and 2D imagery, examine the methods that were used, and survey the areas in which they were applied.