2013
Ramos-Romero, Francisco; Chover, Miguel; Ripolles, Oscar
A multiresolution approach to render 3D models Journal Article
In: Informatica (Netherlands), vol. 24, no. 4, pp. 603-618, 2013, ISSN: 0868-4952, (IF: 0.901, Q2).
Abstract | Links | BibTeX | Tags: 3D, GPU, mesh refinement
@article{RamosRomero2013,
title = {A multiresolution approach to render 3D models},
author = { Francisco Ramos-Romero and Miguel Chover and Oscar Ripolles},
url = {http://hdl.handle.net/10234/160304},
issn = {0868-4952},
year = {2013},
date = {2013-01-01},
journal = {Informatica (Netherlands)},
volume = {24},
number = {4},
pages = {603-618},
publisher = {IOS Press},
abstract = {Image synthesis techniques are present in a wide range of applications as they leverage the amount of information required for creating realistic visualizations. For fast hardware rendering they usually employ a triangle-based representation describing the geometry of the scene. In this paper, we introduce a new and simple framework for performing on-the-fly refinement and simplification of meshes completely on the GPU. As we aim at making easy the integration of level-of-detail management into the creation workflow of artists, the presented method is easy to be implemented. We only need a coarse mesh, its displacement map and a geometry shader. At rendering time, we employ a geometry shader to parallelize the tessellation and displacement steps. The tessellation step performs uniform refinement or simplification operations by applying a fixed subdivision criterion. Our method also exploits coherence by taking advantage of the last computed mesh. We provide a method which offers a flexible integration with standard 3D tools, easy to be implemented, coherence exploitation and wholly processed by the GPU.},
note = {IF: 0.901, Q2},
keywords = {3D, GPU, mesh refinement},
pubstate = {published},
tppubtype = {article}
}
Image synthesis techniques are present in a wide range of applications as they leverage the amount of information required for creating realistic visualizations. For fast hardware rendering they usually employ a triangle-based representation describing the geometry of the scene. In this paper, we introduce a new and simple framework for performing on-the-fly refinement and simplification of meshes completely on the GPU. As we aim at making easy the integration of level-of-detail management into the creation workflow of artists, the presented method is easy to be implemented. We only need a coarse mesh, its displacement map and a geometry shader. At rendering time, we employ a geometry shader to parallelize the tessellation and displacement steps. The tessellation step performs uniform refinement or simplification operations by applying a fixed subdivision criterion. Our method also exploits coherence by taking advantage of the last computed mesh. We provide a method which offers a flexible integration with standard 3D tools, easy to be implemented, coherence exploitation and wholly processed by the GPU.