<link rel="stylesheet" href="styles.cf390c6bfbe7555f.css"> Skip to main content

Vector-field-based deformations for three-dimensional texture synthesis

dc.citation.epage398en_US
dc.citation.issue4en_US
dc.citation.volume6en_US
dc.citation.woscount0
dc.contributor.authorSu, Y. -L.en_US
dc.contributor.authorChang, C. -C.en_US
dc.contributor.authorShih, Z. -C.en_US
dc.contributor.authorTai, W. -K.en_US
dc.contributor.department多媒體工程研究所zh_TW
dc.contributor.departmentInstitute of Multimedia Engineeringen_US
dc.date.accessioned2014-12-08T15:23:12Z
dc.date.available2014-12-08T15:23:12Z
dc.date.issued2012-06-01en_US
dc.description.abstractA major drawback of conventional approaches to three-dimensional (3D) texture synthesis is the lack of fine-scale deformations in the synthesised results. This study presents a novel vector-field-based deformation approach to synthesising 3D textures with fine-scale deformations. The pre-process of this approach constructs feature vectors and similarity sets of voxels from an input 3D exemplar to accelerate neighbourhood matching. The synthesis process first introduces 3D vector fields for deforming synthesised results. A 3D pyramid synthesis technique integrated with 3D vector fields is then used to synthesise 3D textures. The proposed approach uses only eight neighbourhoods of each voxel for neighbourhood matching. Experimental results show that the proposed approach efficiently synthesises 3D textures with fine-scale deformations.en_US
dc.identifier.issn1751-9659en_US
dc.identifier.journalIET IMAGE PROCESSINGen_US
dc.identifier.urihttps://ir.lib.nycu.edu.tw/handle/11536/16287
dc.identifier.wosnumberWOS:000304244400011
dc.language.isoen_USen_US
dc.titleVector-field-based deformations for three-dimensional texture synthesisen_US
dc.typeArticleen_US

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
000304244400011.pdf
Size:
1.21 MB
Format:
Adobe Portable Document Format

License bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed to upon submission
Description: