EUVIMEDEUVIMED entdecken
Crossref

Smooth normals with blobs for surfaces from 3D binary images

Cinthya Ceja, Caleb Rascon, Edgar Garduño, Bruno Carvalho, Gabor Herman

Topology Proceedings · 2022

DOI: 10.25071/1qrs6337 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Topology Proceedings, 2022. Verfasst von Cinthya Ceja und 4 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

One way of representing a real object on a computer screen is by rendering a polygonal mesh, extracted from the discretized version of the object. This discretization is typically a collection of abutting small cubic voxels. In this work, we present a method that generates better renderings of rectangular meshes (with desirable mathematical properties but blocky renderings) created by a boundary detection algorithm. We achieve this by assigning appropriate normals to the vertices of the mesh and taking advantage of how standard computer graphics methods render images. We assign the normals by evaluating, at the vertices of the mesh, the gradient of a linear combination of Kaiser-Bessel-based basis functions that are spherically symmetric. By using signal processing principles, we select the parameters of these functions and assign normals that yield renderings that show smoother surfaces than the renderings produced by the original voxel based models, without modifying the mesh geometry.

Zitieren

APA

Ceja, C., Rascon, C., Garduño, E., Carvalho, B., & Herman, G. (2022). Smooth normals with blobs for surfaces from 3D binary images. Topology Proceedings. https://doi.org/10.25071/1qrs6337

Vancouver

Ceja C, Rascon C, Garduño E, Carvalho B, Herman G. Smooth normals with blobs for surfaces from 3D binary images. Topology Proceedings. 2022. doi:10.25071/1qrs6337

BibTeX
@article{Ceja2022,
  author = {Ceja, Cinthya and Rascon, Caleb and Garduño, Edgar and Carvalho, Bruno and Herman, Gabor},
  title = {Smooth normals with blobs for surfaces from 3D binary images},
  journal = {Topology Proceedings},
  year = {2022},
  doi = {10.25071/1qrs6337}
}
Kostenlos starten →