Takehiko NAGAKURA / Woong-ki SUNG (Department of Architecture, Massachusetts Institute of Technology, Cambridge, USA) Keywords: Augmented Reality, Architecture, Exhibition, Spatial Representation, Design Media Abstract: This paper introduces an application of Augmented Reality (AR) technology for presenting large and complex spatial designs, and discusses its benefit for museum exhibits especially with physical models. We assembled a low-cost prototype that uses common handheld devices and widely-used marker-based sensing method. Architecture and urban forms are difficult to present in museums. A building usually is too big to fit in an exhibition room, and removing architecture from its site to display in a museum would lose some essence since its design is strongly tied with surrounding context. Alternatively, use of traditional scale models, photographs, and drawings is a way of translating the original into useful representations commonly deployed by professional architects. But these methods fragment the building into isolated forms of different media and projections, and ordinary audience is often left clueless about the original architecture. For instance, tests show it is difficult to relate a section with a scale model of a building, or to locate the viewing position of a photograph on the floor plan. What are almost lost in this translation and cause difficulties are the spatial relationships amongst architecture and various forms of its representations. To resolve this situation, we built Multirama, an AR-based prototype, and loaded it with digital media contents about a Renaissance villa designed by Andrea Palladio. On a table, the installation exhibits its 3D-printed partial model, and audience use tablet computers for viewing it with selectively superimposed augmentation, which includes a photogrammetric model sampled through a fieldwork on the building site, geometric models illustrating the foundation and roof tectonics, as well as plan, section and elevation drawings depicted in Palladio’s canon, The Four Books of Architecture. This system capable of simple synchronization of views demonstrates an effective, intuitive means to help audience minds integrate diverse forms of spatial...

Takehiko NAGAKURA / Daniel TSAI / Howard BURNS (Department of Architecture, Massachusetts Institute of Technology, Cambridge, USA) Keywords: Palladio, Architectural Model, Reality Capture, Photogrammetry, UAV Abstract: This paper describes our lab’s experiments with digital capture equipment and techniques used to study the architecture of Andrea Palladio (1508-1580). One initial question is whether widely available technology can be used to create a snapshot of an entire building in terms of geometry, texture, and illumination — and what this ultimately means for the study, teaching, and conservation of architecture. Since spring of 2013, multiple variations of equipment and techniques were tested and compared. The tests include (1) unmanned aerial vehicle (UAVs): low-cost stock drone versus custom-built equipment; (2) photogrammetry: cloud-based services versus local graphics computing solutions; (3) capture method: photogrammetry versus RGBD (Kinect) versus laser scanning techniques, and (4) panoramic video: single camera with reflector versus multiple-camera rig. The technical, comparative findings are described in terms of cost, processing time, and for use cases such as exterior facades, interior spaces, frescoes, and architectural details. Evaluation of the experiments eventually altered our approach, away from an instant snapshot of an entire building, toward a composite, heterogeneous representation: e.g., UAVs capturing aerial views, high resolution photogrammetry for architectural details, and RGBD for tight staircases. This direction seems to be a more realistic and exciting contribution. Information gathered over time, from different digital capture techniques and various people can be combined into a dynamic building database. Such a building database may cover the building studied at various scales, the interior and exterior, the context and site, with and without furniture, in various weather and lighting conditions. As better versions of representations are created, the parts can be compared and selected. Our ongoing research includes development of an online implementation of this idea, which may be open to public input, including crowd-sourced digital capture, as well as interactive visualization and navigation of the building database via game-engine...

Margot BERNSTEIN (Columbia University, New York, USA) Keywords: Session: The State of 3D Modeling of Cultural Heritage in the Age of Augmented Reality, Xbox Kinect, UAVs, and the Oculus Rift Abstract: Museums around the world have begun to question whether they can or should maintain their period rooms. This paper asks how digital technologies can preserve period rooms and transform our knowledge of their contents by digitally re-imagining one of the great eighteenth-century French period rooms in the Wrightsman Galleries at the Metropolitan Museum of Art in New York. In virtual period rooms, visitors are invited to engage with the objects and people that occupied and enlivened these interiors. High resolution details and interactive, rotational 3D models of eighteenth-century objects—be they pieces of furniture, lighting fixtures, or dinnerware fashioned from native or exotic woods, an assortment of fabrics, powder pink porcelain, lacquer, and more—offer unprecedented insight into the creation, decoration, and display of the first modern domestic interiors. Virtual period rooms, moreover, can be digitally populated with the artisans who furnished such spaces, the wealthy people who inhabited them, and the servants who served the wealthy. Video clips from period films, a recent digitized model of an eighteenth-century game table created and presented to the public by the Metropolitan Museum, and high resolution photographs of objects whose complicated configurations and lavish decoration are either hidden from or inaccessible to museum visitors, all have the potential to re-present period rooms as never...

Margot BERNSTEIN (Columbia University, New York, USA) Keywords: Session: The State of 3D Modeling of Cultural Heritage in the Age of Augmented Reality, Xbox Kinect, UAVs, and the Oculus Rift Abstract: My paper emphasizes the manner and extent to which twenty-first-century technologies can help us to understand the mobility and meaning of eighteenth-century furniture. It argues not only for a rethinking of decorative arts objects, but also for a rethinking of (indeed, a revolution in) scholarly approaches to these objects—approaches that move beyond the written word and photographic image and embrace the video and digital technologies about which eighteenth-century society could not have dreamed, and of which twenty-first-century art historians have yet to take full advantage. The paper centers on a folding card table of c.1755-65 attributed to Bernard II van Risenburgh in the collection of the Metropolitan Museum of Art in its capacity to function not only as a window into eighteenth-century French cultural practice, but also as an active participant within that practice. This table embodies secrecy, chance, and constant flux, each of which was germane to the card games that were played on its surface; it expresses some of the most subtle and characteristic aspects of eighteenth-century culture. Yet, access to and insight into the van Risenburgh table’s appearance, functionality, and broader socio-cultural significance are limited for visitors in the Metropolitan’s gallery, where this piece of furniture remains both out of reach and stationary. Because the table has only been captured in still photographs, audiences viewing this object on the museum’s website are similarly deprived of a comprehensive understanding of its dynamic, kinetic existence. In the absence of a digital model of the card table, I rely upon a digitized recreation of an eighteenth-century game table by David Roentgen that was created and made publicly available by the Metropolitan Museum on the occasion of a recent exhibition in my efforts to illustrate digital technology’s crucial role in illuminating eighteenth-century furniture’s capacities to direct, reflect, and manifest...

Chairs: Bernhard FRISCHER, USA / Gabriele GUIDI, Italy The purpose of this session is to take stock of the current state of the art of 3D modeling of cultural heritage objects of all scales (from a vase or statue, to a building or an entire landscape). It seems particularly opportune to do so in light of the dramatically lower costs of devices for 3D data capture and display. Papers fitting one or more of the following three descriptions would be especially welcome: (1) cultural heritage projects exploiting Augmented Reality and/or immersive, 3D display devices such as the Oculus Rift; (2) projects collecting 3D data using low-cost devices and methods such as SfM, Kinect or Unmanned Aerial Vehicles;  (3) 3D modeling projects (especially those involving reconstructions) that have as their goal not simply illustrating previously existing knowledge of the past but serving as tools to see or understand features of the past that can only emerge after we have made the 3D model. Papers should concentrate less on the “how” of 3D modeling (fairly well understood by now) than on the “why” (i.e., what is the scientific gain in knowledge that results from applying the new technology?). Topics falling into category (1) should, whenever possible, provide the results of summative assessment: we are interested not simply in bright ideas and clever demonstrations but in proof of concept or full-scale deployment. For example, if a  claim is made that AR can help promote better public understanding of cultural heritage, did the pilot AR project actually produce measurably positive results? Topics falling into category (2) should ideally include a comparison of the results of using low-cost and high-end data gathering devices. What are the strengths and weaknesses of the low-cost approach? Is a low-cost device good for certain cultural heritage applications but not others? For example, we welcome a paper comparing the resolution and accuracy of 3D meshes of a statue resulting from a SfM approach vs. one resulting from traditional scanning....