• 综合
  • 标题
  • 关键词
  • 摘要
  • 学者
  • 期刊-刊名
  • 期刊-ISSN
  • 会议名称
搜索

作者:

Li, Yan (Li, Yan.) | Yin, Baocai (Yin, Baocai.) (学者:尹宝才) | Kong, Dehui (Kong, Dehui.) (学者:孔德慧)

收录:

EI Scopus

摘要:

To avoid noticeable seams between texture patches and minimize the distortion of the pattern when tiling a texture on surfaces, several texture synthesis approaches have been proposed in the past few years, among which patch-based texture synthesis approaches have achieved better effect and higher speed compared with the point-based texture synthesis approaches. However, as a typical patch-based texture synthesis approach, Efros patch-based texture synthesis approach is still not real-time. With the help of 'texture consistency', this paper presents a novel approach for texture synthesis to match patch along a spiral path. The algorithm searches for new matching patch among the neighborhoods of the previous matched patch along a spiral path and continues this process in a recursive manner until the synthesizing image is fully filled. Compared with the Efros approach under the same condition of synthetic quality, the approach presented in this paper is 10 times faster on average. Experiments show that synthesizing results are quite satisfactory.

关键词:

Algorithms Image analysis Image matching Image quality Pattern matching Surfaces Textures

作者机构:

  • [ 1 ] [Li, Yan]Key Laboratory of Multimedia and Intelligent Software, Beijing University of Technology, Beijing 100022, China
  • [ 2 ] [Li, Yan]Guangxi Jiaotong Vocational and Technical College, Nanning 530023, China
  • [ 3 ] [Yin, Baocai]Key Laboratory of Multimedia and Intelligent Software, Beijing University of Technology, Beijing 100022, China
  • [ 4 ] [Kong, Dehui]Key Laboratory of Multimedia and Intelligent Software, Beijing University of Technology, Beijing 100022, China

通讯作者信息:

电子邮件地址:

查看成果更多字段

相关关键词:

相关文章:

来源 :

Journal of Information and Computational Science

ISSN: 1548-7741

年份: 2007

期: 1

卷: 4

页码: 107-113

被引次数:

WoS核心集被引频次: 0

SCOPUS被引频次:

ESI高被引论文在榜: 0 展开所有

万方被引频次:

中文被引频次:

近30日浏览量: 2

在线人数/总访问数:113/3601943
地址:北京工业大学图书馆(北京市朝阳区平乐园100号 邮编:100124) 联系我们:010-67392185
版权所有:北京工业大学图书馆 站点建设与维护:北京爱琴海乐之技术有限公司