大型结构的纹理合成优化

Xiuxia Cai, Haoyu Li, Sandong Guo, Huimin Du
{"title":"大型结构的纹理合成优化","authors":"Xiuxia Cai, Haoyu Li, Sandong Guo, Huimin Du","doi":"10.1109/ICTC51749.2021.9441501","DOIUrl":null,"url":null,"abstract":"Exemplar-based texture synthesis methods aim to generate arbitrarily larger textures from limited exemplars without strong artifacts or introducing obvious unnatural visual elements. While existing non-parametric exemplar-based synthesis methods have made remarkable progress on reduce artifacts. Most of them have not been good enough to avoid introducing error-mixed larger scale in synthesis textures. In addition, in noisy large-scale regular texture, there always exists the large-scale motif texture. These large-scale motifs are difficult to correct synthesized for texture synthesis. So in this paper we focus on the problem of large scale structures synthesized problem in texture synthesis. In order to avoid introducing error large scale in texture synthesis, diamond searching with decreasing radius method is proposed in the block search step. The patch distance measure is designed by considering local principal curvature. Different from most of the existing techniques, no extra guided channel is needed. The exemplar contends large-scale regular structure will be extracted a motif at first. Then the motif will be synthesized by patch-based method. The experiment results show that the proposed method achieves significant improvements.","PeriodicalId":352596,"journal":{"name":"2021 2nd Information Communication Technologies Conference (ICTC)","volume":"64 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-05-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Texture Synthesis Optimization for Large Scale Structures\",\"authors\":\"Xiuxia Cai, Haoyu Li, Sandong Guo, Huimin Du\",\"doi\":\"10.1109/ICTC51749.2021.9441501\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Exemplar-based texture synthesis methods aim to generate arbitrarily larger textures from limited exemplars without strong artifacts or introducing obvious unnatural visual elements. While existing non-parametric exemplar-based synthesis methods have made remarkable progress on reduce artifacts. Most of them have not been good enough to avoid introducing error-mixed larger scale in synthesis textures. In addition, in noisy large-scale regular texture, there always exists the large-scale motif texture. These large-scale motifs are difficult to correct synthesized for texture synthesis. So in this paper we focus on the problem of large scale structures synthesized problem in texture synthesis. In order to avoid introducing error large scale in texture synthesis, diamond searching with decreasing radius method is proposed in the block search step. The patch distance measure is designed by considering local principal curvature. Different from most of the existing techniques, no extra guided channel is needed. The exemplar contends large-scale regular structure will be extracted a motif at first. Then the motif will be synthesized by patch-based method. The experiment results show that the proposed method achieves significant improvements.\",\"PeriodicalId\":352596,\"journal\":{\"name\":\"2021 2nd Information Communication Technologies Conference (ICTC)\",\"volume\":\"64 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2021-05-07\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2021 2nd Information Communication Technologies Conference (ICTC)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICTC51749.2021.9441501\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 2nd Information Communication Technologies Conference (ICTC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICTC51749.2021.9441501","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

基于样例的纹理合成方法旨在从有限的样例中生成任意较大的纹理,而不需要强人工制品或引入明显的非自然视觉元素。而现有的基于非参数样例的合成方法在减少伪影方面取得了显著的进展。它们中的大多数都不够好,无法避免在合成纹理中引入更大尺度的误差混合。此外,在嘈杂的大尺度规则织构中,总会存在大尺度母题织构。这些大规模的图案在纹理合成中很难进行校正。因此,本文主要研究纹理合成中大规模结构的合成问题。为了避免在纹理合成中引入大规模误差,在块搜索步骤中提出了半径递减的菱形搜索方法。考虑局部主曲率,设计了斑块距离测度。与大多数现有技术不同,它不需要额外的引导通道。范例认为,大型规则结构将首先被提取为一个母题。然后采用基于patch的方法合成motif。实验结果表明,该方法取得了显著的改进效果。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Texture Synthesis Optimization for Large Scale Structures
Exemplar-based texture synthesis methods aim to generate arbitrarily larger textures from limited exemplars without strong artifacts or introducing obvious unnatural visual elements. While existing non-parametric exemplar-based synthesis methods have made remarkable progress on reduce artifacts. Most of them have not been good enough to avoid introducing error-mixed larger scale in synthesis textures. In addition, in noisy large-scale regular texture, there always exists the large-scale motif texture. These large-scale motifs are difficult to correct synthesized for texture synthesis. So in this paper we focus on the problem of large scale structures synthesized problem in texture synthesis. In order to avoid introducing error large scale in texture synthesis, diamond searching with decreasing radius method is proposed in the block search step. The patch distance measure is designed by considering local principal curvature. Different from most of the existing techniques, no extra guided channel is needed. The exemplar contends large-scale regular structure will be extracted a motif at first. Then the motif will be synthesized by patch-based method. The experiment results show that the proposed method achieves significant improvements.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信