Guest Introduction The Changing Shape of Computer Vision in the Twenty-First Century.pdfVIP

下载本文档

6
0
约3.6万字
约 8页
2017-04-11 发布于江苏
举报
版权申诉

Guest Introduction The Changing Shape of Computer Vision in the Twenty-First Century.pdf

1、原创力文档（book118）网站文档一经付费（服务费），不意味着购买了该文档的版权，仅供个人/单位学习、研究之用，不得用于商业用途，未经授权，严禁复制、发行、汇编、翻译或者网络传播等，侵权必究。。
2、本站所有内容均由合作方或网友上传，本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺！文档内容仅供研究参考，付费前请自行鉴别。如您付费，意味着您自己接受本站规则且自行承担风险，本站不退款、不进行额外附加服务；查看《如何避免下载的几个坑》。如果您已付费下载过本站文档，您可以点击这里二次下载。
3、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等，请点击“版权申诉”（推荐），也可以打举报电话：400-050-0827(电话支持时间：9:00-18:30)。
4、该文档为VIP文档，如果想要下载，成为VIP会员后，下载免费。
5、成为VIP后，下载本文档将扣除1次下载权益。下载后，不支持退款、换文档。如有疑问请联系我们。
6、成为VIP后，您将拥有八大权益，权益包括：VIP文档下载权益、阅读免打扰、文档格式转换、高级专利检索、专属身份标志、高级客服、多端互通、版权登记。
7、VIP文档为合作方或网友上传，每下载1次，网站将根据用户上传文档的质量评分、类型等，对文档贡献者给予高额补贴、流量扶持。如果你也想贡献VIP文档。上传文档

Guest Introduction The Changing Shape of Computer Vision in the Twenty-First Century

International Journal of Computer Vision 50(2), 103–110, 2002 c? 2002 Kluwer Academic Publishers. Manufactured in The Netherlands. Guest Introduction: The Changing Shape of Computer Vision in the Twenty-First Century MUBARAK SHAH Computer Vision Lab, School of Electrical Engineering and Computer Science, University of Central Florida, Orlando, FL 32816, USA shah@cs.ucf.edu 1. Introduction Computer vision started as an Artificial Intelligence (AI) problem. For this reason, vision has also been called image understanding. The original goal of vi- sion was to understand a single image of a scene, lo- cate and identify objects, determine their structures, spatial arrangements, relationship with other objects, etc. The MIT copy demo was a good example of this. The idea in the copy demo was roughly like this: to have a computer vision program analyze an image of a scene containing several stacked blocks, recover the structure of the blocks, and generate a script, for a robot to build an exact copy of the block structure.1 This was actually a high level vision problem. This copy demo was one of the motivations for the work in the blocks world like consistent line labeling, analysis of polyhe- dral junctions, etc. The researchers soon realized that low level vision was not robust enough; they were not even able to extract lines from images to be used in this work. Therefore, it became necessary to first solve low level vision problems (see for example Binford-Horn line finder (Horn, 1973), before the high level vision problems could be attacked. Research in low level vi- sion continued for some time. During the seventies, Marr (1982) captured the at- tention of vision researchers. He popularized shape- from-X, and the use of ideas from the human vision system, among other things. Since one dimension is lost during the projection of a 3D world onto 2D im- ages, the aim of shape-from-X methods is to recover that lost dimension. The next two or three decades were spent develop