142 research outputs found
Shape from Distortion: 3D Range Scanning of Mirroring Objects
No full textObjects with mirroring surfaces are left out of the scope of most recent 3D scanning methods. We developed a new acquisition approach, shape-from-distortion, that focuses on that category of objects, requires only a still camera and a monitor, and generates high quality range scans (plus a normal field). Our contributions are a novel acquisition technique based on environment matting [Chuang et al. 2000] and corresponding geometry reconstruction method that recovers a very precise geometry model for mirroring objects
3D Acquisition of Mirroring Objects using Striped Patterns
No full textObjects with mirroring optical characteristics are left out of the scope of most 3D scanning methods. We present here a new automatic acquisition approach, shape-from-distortion, that focuses on that category of objects, requires only a still camera and a color monitor, and produces range scans (plus a normal and a reflectance map) of the target. Our technique consists of two steps: first, an improved environment matte is captured for the mirroring object, using the interference of patterns with different frequencies to obtain sub-pixel accuracy. Then, the matte is converted into a normal and a depth map by exploiting the self-coherence of a surface when integrating the normal map along different paths. The results show very high accuracy, capturing even smallest surface details. The acquired depth maps can be further processed using standard techniques to produce a complete 3D mesh of the object
3D Data Acquisition - Eurographics 2002 Tutorial Notes
No full text3D scanners and image acquisition systems are rapidly becoming more affordable and allow to build highly accurate models of real 3D objects in a cost- and time-effective manner. This tutorial will present the potential of this technology, review the state of the art in model acquisition methods, and will discuss the 3D acquisition pipeline from physical acquisition until the final digital model. First, different optical scanning techniques (e.g. structured light triangulation, time-of-flight approaches) will briefly be presented. Other acquisition related issues including the design of the scanning studio will be discussed and evaluated. In the area of registration, we will consider both the problems of initially aligning individual scans, and of refining this alignment with variations of the Iterative Closest Point method. For scan integration and mesh reconstruction, we will compare various methods for computing, interpolating and approximating surfaces. We will then look at various ways in which surface properties such as color and reflectance can be extracted from acquired imagery. Finally, we will examine techniques for the efficient management and rendering of very large, attribute-rich meshes, including methods for the construction of simplified triangle-based representation and sample-based rendering approaches
topicmodels: An R Package for Fitting Topic Models
No full textTopic models allow the probabilistic modeling of term frequency occurrences in documents. The fitted model can be used to estimate the similarity between documents as well as between a set of specified keywords using an additional layer of latent variables which are referred to as topics. The R package topicmodels provides basic infrastructure for fitting topic models based on data structures from the text mining package tm. The package includes interfaces to two algorithms for fitting topic models: the variational expectation-maximization algorithm provided by David M. Blei and co-authors and an algorithm using Gibbs sampling by Xuan-Hieu Phan and co-authors.
topicmodels: An R Package for Fitting Topic Models
No full textTopic models allow the probabilistic modeling of term frequency occurrences in documents.
The fitted model can be used to estimate the similarity between documents as
well as between a set of specified keywords using an additional layer of latent variables
which are referred to as topics. The R package topicmodels provides basic infrastructure
for fitting topic models based on data structures from the text mining package tm. The
package includes interfaces to two algorithms for fitting topic models: the variational
expectation-maximization algorithm provided by David M. Blei and co-authors and an
algorithm using Gibbs sampling by Xuan-Hieu Phan and co-authors
Digital Collections of Real World Objects
No full textReal world objects, such as works of art, archeological artifacts and even common everyday objects, exhibit large variations in color due to the way light is reflected from their surfaces. A high quality digitization method must be capable of capturing these effects if the digital models generated from the real objects are to look realistic.In this article, we present an efficient method for acquiring high quality models of real world objects. The resulting digital models can be viewed under arbitrary viewing and lighting conditions. The efficient acquisition technique, small size, high quality, and versatility of the generated models make this technique well suited for large digital collections
7th ACM Symposium on Solid Modeling and Applications; Tutorial T2: A Framework for the Acquisition, Processing and Interactive Display of High Quality 3D Models
No full textMulti-View Stereo for Community Photo Collections
No full textWe present a multi-view stereo algorithm that addresses the extreme changes in lighting, scale, clutter, and other effects in large online community photo collections. Our idea is to intelligently choose images to match, both at a per-view and per-pixel level. We show that such adaptive view selection enables robust performance even with dramatic appearance variability. The stereo matching technique takes as input sparse 3D points reconstructed from structure-from-motion methods and iteratively grows surfaces from these points. Optimizing for surface normals within a photoconsistency measure significantly improves the matching results. While the focus of our approach is to estimate high-quality depth maps, we also show examples of merging the resulting depth maps into compelling scene reconstructions. We demonstrate our algorithm on standard multi-view stereo datasets and on casually acquired photo collections of famous scenes gathered from the Internet
New acquisition techniques for real objects and light sources in computer graphics
No full textAccurate representations of objects and light sources in a scene model are a crucial prerequisite for realistic image synthesis using computer graphics techniques. This thesis presents techniques for the effcient acquisition of real world objects and real world light sources, as well as an assessment of the quality of the acquired models. Making use of color management techniques, we setup an appearance reproduction pipeline that ensures best-possible reproduction of local light reflection with the available input and output devices. We introduce a hierarchical model for the subsurface light transport in translucent objects, derive an acquisition methodology, and acquire models of several translucent objects that can be rendered interactively. Since geometry models of real world objects are often acquired using 3D range scanners, we also present a method based on the concept of modulation transfer functions to evaluate their accuracy. In order to illuminate a scene with realistic light sources, we propose a method to acquire a model of the near-field emission pattern of a light source with optical prefiltering. We apply this method to several light sources with different emission characteristics and demonstrate the integration of the acquired models into both, global illumination as well as hardware-accelerated rendering systems.Exakte Repräsentationen der Objekte und Lichtquellen in einem Modell einer
Szene sind eine unerlässliche Voraussetzung für die realistische Bilderzeugung
mit Techniken der Computergraphik. Diese Dissertation beschäftigt sich mit der
effizienten Digitalisierung von realen Objekten und realen Lichtquellen. Dabei
werden sowohl neue Digitalisierungstechniken als auch Methoden zur Bestimmung der Qualität der erzeugten Modelle vorgestellt. Wir schlagen eine Verarbeitungskette zur Digitalisierung und Wiedergabe der Farbe und Spekularität von Objekten vor, die durch Ausnutzung von Farbmanagementtechniken eine bestmögliche Wiedergabe des Objekts unter Verwendung der gegebenen Ein- und Ausgabegeräte ermöglicht. Wir führen weiterhin ein hierarchisches Modell für den Lichttransport im Inneren von Objekten aus durchscheinenden Materialien sowie eine zugehörige Akquisitionsmethode ein und digitalisieren mehrere reale Objekte. Die dabei erzeugten Modelle können in Echtzeit angezeigt werden. Die Geometrie realer Objekte spielt eine entscheidende Rolle in vielen Anwendungen und wird oftmals unter Verwendung von 3D Scannern digitalisiert. Wir entwickeln daher eine Methode zur Bestimmung der Genauigkeit eines 3D Scanners, die auf dem Konzept der Modulationstransferfunktion basiert. Um eine Szene mit realen Lichtquellen beleuchten zu können, schlagen wir ferner eine Methode zur Erfassung der Nahfeldabstrahlung eine Lichtquelle vor, bei der vor der Digitalisierung ein optischer Filterungsschritt durchgeführt wird.
Wir wenden diese Methode zur Digitalisierung mehrerer Lichtquellen mit unterschiedlichen Abstrahlcharakteristika an und zeigen auf, wie die dabei erzeugten Modelle in globalen Beleuchtungsberechnungen sowie bei der Bildsynthese mittels moderner Graphikkarten verwendet werden können
- …
