Intuition on boundaries, shocks and smoothing with PDEs

J. Tumblin
{"title":"Intuition on boundaries, shocks and smoothing with PDEs","authors":"J. Tumblin","doi":"10.1145/1281500.1281603","DOIUrl":null,"url":null,"abstract":"When drawing or painting, many artists capture visual appearance with a coarse-to-fine sequence of boundaries and shadings. Many begin with a sketch of large, important scene features and then gradually add finer, more subtle details. Initial sketches of a painting may hold sharply defined boundaries around large, smoothly shaded regions for the largest, highest contrast and most important scene features. The artist then adds more shadings and boundaries to build up fine details, to fill in the visually empty regions and to capture rich detail everywhere. Such a procedure implicitly defines a hierarchical or coarse-to-fine scene decomposition, and we can use edge preserving filter operations such as the bilateral filters and PDEs to construct a similar decomposition for detail-preserving contrast reduction. This detail hierarchy is based on scene boundaries and shadings, an approach markedly different from the bandpass linear filter decompositions (e.g. image pyramids) that may suffer from halo artifacts. A suitable method for contrast reduction will produce no such halos.","PeriodicalId":184610,"journal":{"name":"ACM SIGGRAPH 2007 courses","volume":"56 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2007-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"ACM SIGGRAPH 2007 courses","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/1281500.1281603","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

When drawing or painting, many artists capture visual appearance with a coarse-to-fine sequence of boundaries and shadings. Many begin with a sketch of large, important scene features and then gradually add finer, more subtle details. Initial sketches of a painting may hold sharply defined boundaries around large, smoothly shaded regions for the largest, highest contrast and most important scene features. The artist then adds more shadings and boundaries to build up fine details, to fill in the visually empty regions and to capture rich detail everywhere. Such a procedure implicitly defines a hierarchical or coarse-to-fine scene decomposition, and we can use edge preserving filter operations such as the bilateral filters and PDEs to construct a similar decomposition for detail-preserving contrast reduction. This detail hierarchy is based on scene boundaries and shadings, an approach markedly different from the bandpass linear filter decompositions (e.g. image pyramids) that may suffer from halo artifacts. A suitable method for contrast reduction will produce no such halos.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
直觉的边界,冲击和平滑与pde
当绘画或绘画时,许多艺术家用粗到细的边界和阴影序列来捕捉视觉外观。许多人从大的、重要的场景特征的草图开始,然后逐渐添加更精细、更微妙的细节。一幅画的初始草图可能会在大的、平滑的阴影区域周围有明确的边界,以获得最大、最高的对比度和最重要的场景特征。然后,艺术家添加更多的阴影和边界,以建立精细的细节,填补视觉空白区域,并捕捉丰富的细节无处不在。这样的过程隐式地定义了分层或粗到细的场景分解,我们可以使用边缘保持滤波器操作,如双边滤波器和偏微分方程来构建类似的分解,以保持细节的对比度降低。这种细节层次结构是基于场景边界和阴影,这是一种明显不同于带通线性滤波器分解(例如图像金字塔)的方法,可能会受到光晕伪影的影响。一种合适的对比度降低方法将不会产生这样的光晕。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Temporal coherence Session details: Course 24: GPGPU: general-purpose cmputation on graphics hardware Session details: Course 11: Practical least-suqares for computer graphics The mobile 3D ecosystem A CMOS camera-based man-machine input device for large-format interactive displays
×
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
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 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 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1