Search

Semantic segmentation is a widespread image analysis task; in some applications, it requires such high accuracy that it still has to be done manually, taking a long time. Deep learning-based approaches can significantly reduce such times, but current automated solutions may produce results below expert standards. We propose agLab, an interactive tool for the rapid labelling and analysis of orthoimages that speeds up semantic segmentation. TagLab follows a human-centered artificial intelligence approach that, by integrating multiple degrees of automation, empowers human capabilities. We evaluated TagLab's efficiency in annotation time and accuracy through a user study based on a highly challenging task: the semantic segmentation of coral communities in marine ecology. In the assisted labelling of corals, TagLab increased the annotation speed by approximately 90% for nonexpert annotators while preserving the labelling accuracy. Furthermore, human-machine interaction has improved the accuracy of fully automatic predictions by about 7% on average and by 14% when the model generalizes poorly. Considering the experience done through the user study, TagLab has been improved, and preliminary investigations suggest a further significant reduction in annotation times.

Annotation support in interactive systems is often considered a simple task by the CG community, since it entails the apparently easy selection of a region and its connection with some information. The reality appears more complex. The scope of this paper is twofold: first, to review the status of this domain, discussing and characterizing several approaches proposed in literature to manage annotations over geometric models; second, to present in detail an innovative solution proposed and assessed in the framework of Cultural Heritage (CH) applications, called ClippingVolumes. At the annotation definition stage ClippingVolumes uses 3D data to characterize the annotation region; subsequently, annotations are visualized by adopting a two-pass rendering solution which uses stencil buffers, thus without introducing new geometric elements, changing the topology or duplicating geometry elements. It solves most of the issues that afflict the current state of the art, such as fragmentation, annotation transfer to multiple representations, and multi-resolution data encoding. The latter is a mandatory requirement to produce efficient web-based systems. We implemented and we fully tested this approach in the framework of a complex system that supports the documentation of CH restoration projects.

Bending-active structures are able to efficiently produce complex curved shapes from flat panels. The desired deformation of the panels derives from the proper selection of their elastic properties. Optimized panels, called FlexMaps, are designed such that, once they are bent and assembled, the resulting static equilibrium configuration matches a desired input 3D shape. The FlexMaps elastic properties are controlled by locally varying spiraling geometric mesostructures, which are optimized in size and shape to match specific bending requests, namely the global curvature of the target shape. The design pipeline starts from a quad mesh representing the input 3D shape, which defines the edge size and the total amount of spirals: every quad will embed one spiral. Then, an optimization algorithm tunes the geometry of the spirals by using a simplified pre-computed rod model. This rod model is derived from a non-linear regression algorithm which approximates the non-linear behavior of solid FEM spiral models subject to hundreds of load combinations. This innovative pipeline has been applied to the project of a lightweight plywood pavilion named FlexMaps Pavilion, which is a single-layer piecewise twisted arch that fits a bounding box of 3.90x3.96x3.25 meters. This case study serves to test the applicability of this methodology at the architectural scale. The structure is validated via FE analyses and the fabrication of the full scale prototype.

Reflectance transformation imaging (RTI) is a computational photography technique widely used in the cultural heritage and material science domains to characterize relieved surfaces. It basically consists of capturing multiple images from a fixed viewpoint with varying lights. Handling the potentially huge amount of information stored in an RTI acquisition that consists typically of 50–100 RGB values per pixel, allowing data exchange, interactive visualization, and material analysis, is not easy. The solution used in practical applications consists of creating “relightable images” by approximating the pixel information with a function of the light direction, encoded with a small number of parameters. This encoding allows the estimation of images relighted from novel, arbitrary lights, with a quality that, however, is not always satisfactory. In this paper, we present NeuralRTI, a framework for pixel-based encoding and relighting of RTI data. Using a simple autoencoder architecture, we show that it is possible to obtain a highly compressed representation that better preserves the original information and provides increased quality of virtual images relighted from novel directions, especially in the case of challenging glossy materials. We also address the problem of validating the relight quality on different surfaces, proposing a specific benchmark, SynthRTI, including image collections synthetically created with physical-based rendering and featuring objects with different materials and geometric complexity. On this dataset and as well on a collection of real acquisitions performed on heterogeneous surfaces, we demonstrate the advantages of the proposed relightable image encoding.

We propose ReviewerNet, an online, interactive visualization system aimed to improve the reviewer selection process in the academic domain. Given a paper submitted for publication, we assume that good candidate reviewers can be chosen among the authors of a small set of pertinent papers; ReviewerNet supports the construction of such set of papers, by visualizing and exploring a literature citation network. The system helps journal editors and Program Committee members to select reviewers that do not have any conflict-of-interest and are representative of different research groups, by visualising the careers and co-authorship relations of candidate reviewers. The system is publicly available, and is demonstrated in the field of Computer Graphics.

[EN] Digital technologies are now mature for producing high quality digital replicas of Cultural Heritage (CH) assets. The research results produced in the last decade ignitedan impressive evolution and consolidation of the technologies for acquiring high-quality digital three-dimensional (3D)models, encompassing both geometry and color. What remains still an open problem is how to deliver those data and related knowledge to our society. The web is nowadays the main channel for the dissemination of knowledge. Emerging commercial solutions for web-publishing of 3D data are consolidating and becoming a de-facto standard for many applications(e-commerce, industrial products, education, etc.).In this framework, CH is a very specific domain, requiring highly flexible solutions. Some recent experiences arepresented, aimed at providing a support to the archival of archaeological3Ddata, supporting web-based publishing of very high-resolution digitization results and finally enabling the documentation of complex restoration actions. All those examples have been recently implemented on the open-source 3D Heritage Online Presenter (3DHOP)platform, developed at CNR-ISTI, [ES] Las tecnologías digitales estánahora maduraspara producir réplicas digitales de alta calidad de valores activos del patrimonio cultural (CH). Los resultados de la investigación producidos en la última década han mostrado una evolución impresionante y una consolidación de las tecnologías para la capturade modelos digitales tridimensionales (3D)de alta calidad, que abarcanla geometríay el color.Lo que queda aún por resolver estárelacionado con la forma de distribuirlos datos y el conocimiento relacionado conla sociedad. La web es hoy en día el principal canal utilizado para divulgarel conocimiento. Las soluciones comerciales nuevas relacionadas con la publicación en la red de datos en 3D se están consolidando y convirtiendo en un estándar de facto para muchas aplicaciones(comercio electrónico, productos industriales, educación, etc.). En este escenario, el patrimonio culturales un dominio muy específico, que requiere soluciones muyflexibles.Se presentan algunas experiencias recientes, destinadasa proporcionar un apoyo al archivo de los datos arqueológicos3D, la publicaciónwebde los resultados de digitalización de muy alta resoluciónque permitenfinalmente la documentación de trabajos de restauracióncomplejos. Todos estos ejemplos se han implementado recientemente en la plataforma 3D Heritage Online Presenter(3DHOP)de código abierto, desarrolladaen el CNR-ISTI., The research leading to these results has received funding from the EU 7th Framework Programme (FP7/2007-2013) under grant agreement no. 654119 (EC "PARTHENOS" project) and EU H2020 Programme (“EMOTIVE: EMOTIve Virtual cultural Experiences through personalized storytelling”, H2020-SC6-CULT-COOP-08-2016) under grant agreement no. 727188.

Close-Range Photogrammetry (CRP) and Reflectance Transformation Imaging (RTI) are two of the most used image-based techniques when documenting and analyzing Cultural Heritage (CH) objects. Nevertheless, their potential impact in supporting study and analysis of conservation status of CH assets is reduced as they remain mostly applied and analyzed separately. This is mostly because we miss easy-to-use tools for of a spatial registration of multimodal data and features for joint visualisation gaps. The aim of this paper is to describe a complete framework for an effective data fusion and to present a user friendly viewer enabling the joint visual analysis of 2D/3D data and RTI images. This contribution is framed by the on-going implementation of automatic multimodal registration (3D, 2D RGB and RTI) into a collaborative web platform (AIOLI) enabling the management of hybrid representations through an intuitive visualization framework and also supporting semantic enrichment through spatialized 2D/3D annotations.

In the last years, the development of compression algorithms for image collections (e.g., photo albums) has become very popular due to the enormous diffusion of digital photographs. Typically, current solutions create an image sequence from images of the photo album to make them suitable for compression using a High Performance Video Coding (HEVC) encoder. In this study, we investigated a different approach to compress a collection of similar images. Our main idea is to exploit the inter- and intra- patch redundancy to compress the entire set of images. In practice, our approach is equivalent to compress the image set with Vector Quantization (VQ) using a global codebook. Our tests show that our clusterization algorithm is effective for a large number of images.

Relightable images have been widely used as a valuable tool in Cultural Heritage (CH) artifacts, including coins, bas-reliefs, paintings, and epigraphs. Reflection Transformation Imaging (RTI), a commonly used type of relightable images, consists of a per-pixel function which encodes the reflection behavior, estimated from a set of digital photographs acquired from a fixed view. Web visualisation tools for RTI images currently require to transmit substantial quantities of data in order to achieve high fidelity renderings. We propose a web-friendly compact representation for RTI images based on a joint interpolation-compression scheme that combines a PCA-based data reduction with a Gaussian Radial Basis Function (RBF) interpolation exhibiting superior performance in terms of quality/size ratio. This approach can be adapted also to other data interpolation schemes, and it is not limited to Gaussian RBF. The rendering part is simple to implement and computationally efficient allowing real-time rendering on low-end devices.

Relightable images have demonstrated to be a valuable tool for the study and the analysis of coins, bas-relief, paintings, and epigraphy in the Cultural Heritage (CH) field. Reflection Transformation Imaging (RTI) are the most diffuse type of relightable images. An RTI image consists in a per-pixel function which encodes the reflection behavior, estimated from a set of digital photographs acquired from a fixed view. Even if web visualization tools for RTI images are available, high fidelity of the relighted images still requires a high amount of data to be transmitted. To overcome this limit, we propose a web-friendly compact representation for RTI images which allows very high quality of the rendered images with a relatively small amount of data required (in the order of 6-9 standard JPEG color images). The proposed approach is based on a joint interpolation-compression scheme that combines a PCA-based data reduction with a Gaussian Radial Basis Function (RBF) interpolation. We will see that the proposed approach can be adapted also to other data interpolation schemes, and it is not limited to Gaussian RBF. The proposed approach has been compared with several techniques, demonstrating its superior performance in terms of quality/size ratio. Additionally, the rendering part is simple to implement and very efficient in terms of computational cost. This allows real-time rendering also on low-end devices.

Restoration is becoming a quite complex process: a large number of internal and external variables co-exist and may impair it. Among these, the large number of professionals involved and the huge amount of documentation produced can heavily affect the quality of the intervention as well as the possibility to have systemic and informed interventions in the future. In particular, a standardized method for storing restoration data and accessing them is still lacking, and the use of new technologies is still limited and/or not scalable. The paper describes the process of designing and testing an information system (IS) based on three-dimensional (3D) data, aimed to support the restoration of Neptune's Fountain in Bologna. In preparation of the restoration, a major effort was carried out to design and implement a web-based IS able to host all of the data produced, to allow the conservation-restoration specialists to interact on-site with an accurate 3D representation of the elements of the fountain, and to directly reference all information and data produced on the geometry of the model. The paper focuses on the challenges and adopted solutions related to the use of 3D models and the data mapping on 3D surfaces in the context of restoration documentation. Highly detailed visualizations of the models, easy navigation, and usable functionalities to add information directly on the 3D model have been achieved by extending the available solutions and by implementing new mechanisms to overcome the limitations of WebGL and remote rendering. Neptune IS’ development has been extensively experimented in a real context of use. Results and knowledge from the experimentation currently represents the basis for evolving Neptune IS into a possible generic and flexible platform for documentation management in the field of restoration and related methodologies.

This paper introduces PhotoCloud, a realtime web-based system which allows end-users to interactively explore and visualize large mixed datasets composed of (up to) several thousands of calibrated 2D photographs joint together with a high-complexity 3D description of the scene. The system is not tailored to any specific data acquisition process, but aims at generality and flexibility. One major focus in our design is scalability in terms of geometric complexity and number of photographs, tackled by means of multiresolution and dynamic hierarchical representation of the data. The dataset is remotely stored and is accessed by the client through a cache system built over a multi-resolution 3D/2D structure. 2D and 3D data are presented to the user in an integrated seamless fashion. % allowing the adoption of either image-driven or 3D-driven navigation controls. %In the former case, an advanced image-browsing interface represents images in meaningful clusters, while the latter is adopted as the main image selection and presentation criteria. Visualization is completed with smooth transitions between 2D and 3D spaces, iconic representations of the images in the 3D space and, finally, by on the fly projections of the selected images on the 3D scene. Navigation of the image dataset is driven by either an advanced image-browsing interface, or a 3D model which is adopted as the main image selection and presentation criteria. Presentation of 2D images is implemented with smooth transitions between the two spaces, adopting clustering and visual presentation techniques to ensure readability of iconic visualization of the image set in the 3D space and, finally, by projecting on the fly the selected image on the 3D scene. As any system paper, the proposed one is an adaptation of techniques already present in literature, or known in the field. The main contribution of this paper consists in their concertation into a multi-resolution system, which effectively allows a high level of integration of generic 2D and 3D datasets. The paper also discusses several implementation choices and the resulting system is made available as an open source project. A preliminary user study conducted with differently skilled users confirm PhotoCloud main contributions in this visualization context.

Limited bandwidth is a strong constraint when efficient transmission of 3D data to Web clients and mobile applications is needed. In this paper we present a novel multi-resolution WebGL based rendering algorithm which combines progressive loading, view-dependent resolution and mesh compression, providing high frame rates and a decoding speed of million of triangles per second in JavaScript. The method is parallelizable and scalable to very large models. The algorithm is based on the local multi-resolution approaches provided by the community, but ad-hoc solutions had to be studied and implemented to provide adequate performances. In particular, a compression mechanism that reached very high compression rate without impact on rendering performance was implemented. Moreover, the data partition strategy was modified in order to be able to load different types of data (i.e. point clouds) and better adapt to the potentials and limitations of web-based rendering.

Efficient transmission of 3D data to Web clients and mobile applications remains a challenge due to limited bandwidth. Most of the research focus in the context of mesh compression has been on improving compression ratio. However, in this context the use of Javascript on the Web and low power CPUS in mobile applications led to critical computational costs. Progressive decoding improves the user experience by providing a simplified version of the model that refines with time, and it's able to mask latency. Current approaches do so at very poor compression rates or at additional computational cost. The need for better performing algorithms is especially evident with this class of methods where Limper [Limper et al. 2013b] demonstrated how decoding time becomes a limiting factor even at moderately low bandwidths. In this paper we present a novel multi-resolution WebGL based rendering algorithm which combines progressive loading, view-dependent resolution and mesh compression, providing high frame rates and a decoding speed of million of triangles per second in Javascript. This method is parallelizable, robust to non-manifold meshes, and scalable to very large models.

3D Heritage Online Presenter (3DHOP) is a framework for the creation of advanced web-based visual presentations of high-resolution 3D content. 3DHOP has been designed to cope with the specific needs of the Cultural Heritage (CH) field. By using multiresolution encoding, it is able to efficiently stream high-resolution 3D models (such as the sampled models usually employed in CH applications); it provides a series of ready-to-use templates and examples tailored for the presentation of CH artifacts; it interconnects the 3D visualization with the rest of the webpage DOM, making it possible to create integrated presentations schemes (3D + multimedia). In its design and development, we paid particular attention to three factors: easiness of use, smooth learning curve and performances. Thanks to its modular nature and a declarative-like setup, it is easy to learn, configure, and customize at different levels, depending on the programming skills of the user. This allows people with different background to always obtain the required power and flexibility from the framework. 3DHOP is written in JavaScript and it is based on the SpiderGL library, which employs the WebGL subset of HTML5, implementing plugin-free 3D rendering on many web browsers. In this paper we present the capabilities and characteristics of the 3DHOP framework, using different examples based on concrete projects.

The program of scientific investigations planned in the framework of the restoration of Michelangelo's David produced several useful guidelines for defining and developing innovative ways to process and visualize 3D data in cultural heritage applications. Our ultimate goal was to include 3D graphics among the tools which can help restorers select the proper restoration procedures for the task at hand and objectively assess restoration results. For this, the David restoration was an ideal test bed to demonstrate the usefulness of digital 3D models and visualization tools in a restoration project. Because a complex set of scientific investigations was planned before and after the restoration intervention, we could try various methodologies to support restorers and scientists with visualization tools based on 3D digital models.

Thanks to the impressive research results produced in the last decade, digital technologies are now mature for producing high-quality digital replicas of Cultural Heritage (CH) artifacts. At the same time, CH practitioners and scholars have also access to a number of technologies that allow distributing and presenting those models to everybody and everywhere by means of a number of communication platforms. The goal of this chapter is to present some recent technologies for supporting the visualization of complex models, by focusing on the requirements of interactive manipulation and visualization of 3D models on the web and on mobile platforms. The section will present some recent experiences where high-quality 3D models have been used in CH research, restoration and conservation. Some open issues in this domain will also be presented and discussed.

Digital technologies are transforming the way cultural heritage researchers, archaeologists, and curators work by providing new ways to collaborate, record excavations, and restore artifacts. The first Web extra is a video that highlights the Cenobium project, a pioneering Web system for presenting medieval cloisters and sculptures. The second video presents the results of a study using digital 3D technologies to assess the apparent shape similarity of a bronze statuette and drawings, to evaluate a possible innovative attribution hypothesis. The third video presents a very complex restoration project of a statue severely damaged by a recent earthquake in central Italy. The project made extensive use of ICT technologies. The fourth video presents a hypothesis of the original location of some terracotta statues over the old temple of Luni during the Roman Etruscan period in Italy. The fifth video was produced for a 2010 exposition on the Roman Empire held in Tokyo and shows the potential of new visual technologies for presenting works of art and supporting storytelling.

Thanks to the WebGL graphics API specification for the JavaScript programming language, the possibility of using the GPU capabilities in a web browser without the need for an ad-hoc plug-in is now coming true. This paper introduces SpiderGL, a JavaScript library for developing 3D graphics web applications. SpiderGL provides data structures and algorithms to ease the use of WebGL, to define and manipulate shapes, to import 3D models in various formats, to handle asynchronous data loading. We show the potential of this novel library with a number of demo applications. Furthermore, we introduce MeShade, a SpiderGL-based web application for shader material editing from within the web browser, which produces all the code needed for embedding interactive 3D model visualization capabilities inside web pages and online repositories.

Fluid simulations typically produce complex threedimensional iso-surfaces whose geometry and topology change over time. The standard way of representing such "dynamic geometry" is by a set of iso-surfaces that are extracted individually at certain time steps. An alternative strategy is to represent the whole sequence as a four-dimensional tetrahedral mesh. The iso-surface at a specific time step can then be computed by intersecting the tetrahedral mesh with a threedimensional hyperplane. This not only allows to animate the surface continuously over time without having to worry about the topological changes, but also enables simplification algorithms to exploit temporal coherence. We show how to interactively render such four-dimensional tetrahedral meshes by improving previous GPU-accelerated techniques and building an out-of-core multi-resolution structure based on quadric-error simplification. As a second application we apply our framework to time-varying surfaces that result from morphing one triangle mesh into another.

The rapid evolution of automatic shape acquisition technologies will make huge amount of sampled 3D data available in the near future. Cul- tural Heritage (CH) domain is one of the ideal fields of application of 3D scanned data, while some issues in the use of those data are: how to visualize at interactive rates and full quality on commodity computers; how to improve visualization ease of use; how to support the integrated visualization of a virtual 3D artwork and the multimedia data which tell its story. We present here the Virtual Inspector tool. The system allows naive users to inspect a very dense 3D model at interactive frame rates on off- the-shelf PC's, presenting the 3D model and all the multimedia data that has been linked to selected points of its surface. A main goal in the design of the system was to provide the user with a very easy and natural inter- action approach, based on a straightforward "point and click" metaphor. Visualization efficiency is obtained, without sacrificing quality, by adopt- ing a state-of-the-art continuous level-of-detail (LOD) representation. Fi- nally, the adoption of XML encoding of the GUI structure and behavior, makes Virtual Inspector a very flexible and configurable system.

The support of advanced Information Technology (IT) to preservation, restoration and documentation of Cultural Heritage is becoming a very important goal for the research community. Michelangelo's David was one of the first applications of 3D scanning technology on a highly popular work of art. The subsequent restoration campaign, started in 2002 and concluded in 2004, was also a milestone for the adoption of modern scientific analysis procedures and IT tools in the framework of a restoration process. One of the focuses in this restoration was also methodological, i.e. to plan and adopt innovative ways to document the restoration process. In this paper we present the results of an integration of different restoration data (2D and 3D datasets) which has been concluded recently. The recent evolution of HW and SW graphics technologies gave us the possibility to interactively visualize an extremely dense 3D model which incorporates the color information provided by two professional photographic campaigns, made before and after the restoration. Moreover, we present the results concerning the mapping, in this case on the 2D media, of the reliefs produced by restorers to assess and document the status of the marble surface before the restoration took place. This result could lead to new and fascinating applications of computer graphics for preservation, restoration and documentation of Cultural Heritage.

The Multi Triangulation framework (MT) is a very general approach for managing adaptive resolution in triangle meshes. The key idea is arranging mesh fragments at different resolution in a Directed Acyclic Graph (DAG) which encodes the dependencies between fragments, thereby encompassing a wide class of multiresolution approaches that use hierarchies or DAGs with predefined topology. On current architectures, the classic MT is however unfit for real-time rendering, since DAG traversal costs vastly dominate raw rendering costs. In this paper, we redesign the MT framework in a GPU friendly fashion, moving its granularity from triangles to precomputed optimized triangle patches. The patches can be conveniently tri-stripped and stored in secondary memory to be loaded on demand, ready to be sent to the GPU using preferential paths. In this manner, central memory only contains the DAG structure and CPU workload becomes negligible. The major contributions of this work are: a new out-of-core multiresolution framework, that, just like the MT, encompasses a wide class of multiresolution structures; a robust and elegant way to build a well conditioned MT DAG by introducing the concept of V -partitions, that can encompass various state of the art multiresolution algorithms; an efficient multithreaded rendering engine and a general subsystem for the external memory processing and simplification of huge meshes.

Refurbishing Michelangelo's David has produced several useful restoration guidelines. These guidelines can help restorers select the proper procedures for the task and, objectively, assess the results. The work also has helped us develop innovative ways to process and visualize 3D data in cultural heritage projects. The David restoration was an ideal test bed. We try various methodologies to support restorers and scientists with visualization tools based on 3D digital models. 3D digital models can be a tool for undertaking specific investigations, or as supporting media for archiving and integrating the restoration-related information.

A new prospective for the study, documentation and presentation of Cultural Heritage is opened by the joint usage of the tools for the automatic reconstruction of digital 3D models and the instruments for interactive 3D visualization. New techniques are available to perform high-resolution sampling of both the shape and the multi-band reflection properties of painted surfaces. The digital 3D models produced are extremely accurate and rich of information, as it has been proved in the experiments run on the Leonardo's Madonna of the Yarnwinder. This paper presents an overview of the techniques needed to build high-quality 3D models from the raw data produced by the scanning devices and describes how to process those models to make them usable in interactive applications. A very critical point is how to integrate the reflection properties with 3D shape models; this integration (based on texture mapping) allows us to produce textured 3D models that allow a joint visualization of shape and color attributes. The interactive visualization tools developed by ISTI-CNR are described, presenting both their features and potential for the visual presentation and analysis of works of art. Examples of the results obtained on the Madonna of the Yarnwinder are presented.

This paper describes an efficient technique for out-of-core management and interactive rendering of planet sized textured terrain surfaces. The technique, called P-Batched Dynamic Adaptive Meshes (P-BDAM), is based on BDAM structure. Data is partitioned into a set of BDAM tiles, each of them constituted by a pair of geometry bintrees of small triangular patches and an associated texture quadtree. Each triangular patch is a general triangulation of points on a displaced triangle. The proposed framework introduces several advances with respect to the state of the art: thanks to a batched host-to-graphics communication model, we outperform current adaptive tessellation solutions in terms of rendering speed; we guarantee overall geometric continuity, exploiting programmable graphics hardware to cope with the accuracy issues introduced by single precision floating points; we exploit a compressed out of core representation and speculative prefetching for hiding disk latency during rendering of out-of-core data; we efficiently construct high quality simplified representations with a novel distributed out of core simplification algorithm working on a standard PC network.

This paper describes an efficient technique for out-of-core rendering and management of large textured terrain surfaces. The technique, called Batched Dynamic Adaptive Meshes (BDAM) , is based on a paired tree structure: a tiled quadtree for texture data and a pair of bintrees of small triangular patches for the geometry. These small patches are TINs and are constructed and optimized off-line with high quality simplification and tristripping algorithms. Hierarchical view frustum culling and view-dependent texture and geometry refinement is performed at each frame through a stateless traversal algorithm. Thanks to the batched CPU/GPU communication model, the proposed technique is not processor intensive and fully harnesses the power of current graphics hardware. Both preprocessing and rendering exploit out-of-core techniques to be fully scalable and to manage large terrain datasets.

We describe an efficient technique for out-of-core construction and accurate view-dependent visualization of very large surface models. The method uses a regular conformal hierarchy of tetrahedra to spatially partition the model. Each tetrahedral cell contains a precomputed simplified version of the original model, represented using cache coherent indexed strips for fast rendering. The representation is constructed during a fine-to-coarse simplification of the surface contained in diamonds (sets of tetrahedral cells sharing their longest edge). The construction preprocess operates out-of-core and parallelizes nicely. Appropriate boundary constraints are introduced in the simplification to ensure that all conforming selective subdivisions of the tetrahedron hierarchy lead to correctly matching surface patches. For each frame at runtime, the hierarchy is traversed coarse-to-fine to select diamonds of the appropriate resolution given the view parameters. The resulting system can interatively render high quality views of out-of-core models of hundreds of millions of triangles at over 40Hz (or 70M triangles/s) on current commodity graphics platforms.

We recently introduced an efficient technique for out-of-core rendering and management of large textured landscapes. The technique, called Batched Dynamic Adaptive Meshes (BDAM), is based on a paired tree structure: a tiled quadtree for texture data and a pair of bintrees of small triangular patches for the geometry. These small patches are TINs that are constructed and optimized off-line with high quality simplification and tristripping algorithms. Hierarchical view frustum culling and view-dependendent texture/geometry refinement is performed at each frame with a stateless traversal algorithm that renders a continuous adaptive terrain surface by assembling out of core data. Thanks to the batched CPU/GPU communication model, the proposed technique is not processor intensive and fully harnesses the power of current graphics hardware. This paper summarizes the method and discusses the results obtained in a virtual flythrough over a textured digital landscape derived from aerial imaging.