Viewpoint selection is an emerging area in computer graphics with applications in fields such as scene exploration, image-based modeling, and volume visualization. In particular, best view selection algorithms are used to obtain the minimum number of views (or images) in order to understand or model an object or scene better. In this article, we present a unified framework for viewpoint selection and mesh saliency based on the definition of an information channel between a set of viewpoints (input) and the set of polygons of an object (output). The mutual information of this channel is shown to be a powerful tool to deal with viewpoint selection, viewpoint stability, object exploration and viewpoint-based saliency. In addition, viewpoint mutual information is extended using saliency as an importance factor, showing how perceptual criteria can be incorporated to our method. Although we use a sphere of viewpoints around an object, our framework is also valid for any set of viewpoints in a closed scene. A number of experiments demonstrate the robustness of our approach and the good behavior of the proposed measures. ACM Reference Format:Feixas, M., Sbert, M., and González, F. 2009. A unified information-theoretic framework for viewpoint selection and mesh saliency.
Figure 1: Filtering with Continuity Mapping (Snake from www.daz3d.com, 25448 triangles, atlas 700 × 3200). In spite of the artist's fine-tuning, seams are visible with padding alone (top). Continuity Mapping makes any multi-chart parameterization seamless (bottom). AbstractIt is well known that multi-chart parameterizations introduce seams over meshes, causing serious problems for applications like texture filtering, relief mapping and simulations in the texture domain.Here we present two techniques, collectively known as Continuity Mapping, that together make any multi-chart parameterization seamless: Traveler's Map is used for solving the spatial discontinuities of multi-chart parameterizations in texture space thanks to a bidirectional mapping between areas outside the charts and the corresponding areas inside; and Sewing the Seams addresses the sampling mismatch at chart boundaries using a set of stitching triangles that are not true geometry, but merely evaluated on a perfragment basis to perform consistent linear interpolation between non-adjacent texel values. Continuity Mapping does not require any modification of the artist-provided textures or models, it is fully automatic, and achieves continuity with small memory and computational costs.
One-dimensional models for extended-duct and perforated-duct mufflers require the introduction of end corrections in order to account for multidimensional effects at the junctions. In this paper, a numerical two-dimensional finite element calculation has been used in order to obtain information on these end corrections. The results have been validated through comparison with experimental measurements performed with a modified version of the impulse method. Then, the influence of the different geometric characteristics of the mufflers on the end correction have been studied. A general correlation in terms of relevant nondimensional parameters is given for extended-duct mufflers, whereas for perforated mufflers a general correlation has not been obtained due to the eventual coupling with other attenuation mechanisms.
A mbient occlusion is a powerful technique that mimics indirect global illumination at a fraction of its cost.1 Researchers introduced obscurances, the first ambient-occlusion technique, in the computer-game context to allow fast editing, and later used it in production rendering.2 To capture a good view of an object, we must first define what a "good view" means. Researchers have proposed several measures to this end, from heuristical measures to information-theoreticbased approaches. Computer graphics researchers have applied viewpoint quality measures in areas such as scene understanding, 3,4 scene exploration, 5 and volume visualization.6,7 However, existing approaches don't exploit all the information that we can infer from captured data, such as how the object's polygon "sees" the viewpoints.We've developed a new ambient occlusion technique based on an information-theoretic framework.8 Essentially, our method computes a weighted visibility from each object polygon to all viewpoints; we then use these visibility values to obtain the information associated with each polygon. So, just as a viewpoint has information about the model's polygons, the polygons gather information on the viewpoints. We therefore have two measures associated with an information channel defined by the set of viewpoints as input and the object's polygons as output, or vice versa. From this polygonal information, we obtain an occlusion map that serves as a classic ambient occlusion technique. Our approach also offers additional applications, including an importance-based viewpoint-selection guide, and a means of enhancing object features and producing nonphotorealistic object visualizations. Overview: Basic conceptsTo set a context for our work, we first review some basic concepts of information theory, viewpoint selection, and ambient occlusion. Information theoryTo introduce the viewpoint channel and an object's associated information, we present here the two most basic measures of information theory: entropy and mutual information.Let X be a finite set and a random variable, taking values x in X with distribution p(x) = Pr [X = x].
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.