期刊文献+

体育视频全景图合成技术 被引量:9

A Panorama Composition Technique of Sports Video
下载PDF
导出
摘要 研究了面向体育视频的全景图合成技术,提出了通过计算前景区域模板和α混合进行前景合成的方法.首先采用MPEG-4校验模型中的Konrad算法和基于Fisher线性判别准则的外点过滤算法进行全局运动参数估计,并用中值法合成背景全景图;然后将前景帧进行仿射投影后与背景求差值,并用中值滤波和形态学方法处理差值图像,得到前景区域模板;最后采用α混合生成前景序列的全景图,并叠加到背景上.由此得到的全景图能同时展现运动的场景和运动员的动作过程. A technique of composing mosaics from sports videos is proposed. First we select means of the temporal median for background composition, then we propose a foreground mix technique based on foreground region extraction and α-composition. To estimate the global motion, Konrad algorithm investigated in the framework of MPEG-4 and an outlier filter based on Fisher discriminant are used. The foreground mask is extracted from margin image by median filter and morphological filter. We add foreground region to background with α-composition. The final panoramas clearly show the full background and vivid actions of athletes.
出处 《计算机辅助设计与图形学学报》 EI CSCD 北大核心 2005年第11期2552-2557,共6页 Journal of Computer-Aided Design & Computer Graphics
基金 国家自然科学基金(60302028 60473002) 北京市自然科学基金重点(4051004) 北京市科技计划(Z0004024040231) 国家体育总局备战2008年奥运会科技攻关(02005)
关键词 体育视频 全景图 前景区域 sports video panoramas mosaic foreground region
  • 相关文献

参考文献7

  • 1Hsu Chiou-Ting, Tsan Yu-Chun. Mosaics of video sequences with moving objects[J]. Signal Processing: Image Communication, 2004, 19(1): 81~98. 被引量:1
  • 2Nicolas H. New methods for dynamic mosaicking[J]. IEEE Transactions on Image Processing, 2001, 10(8): 1239~1251. 被引量:1
  • 3Irani Michal, Anandan P, Hsu Steve. Mosaic based representations of video sequences and their applications[A]. In: Proceedings of the 5th International Conference on Computer Vision (ICCV), Massachusetts, 1995. 605~611. 被引量:1
  • 4Bartoli Adrien, Dalal Navneet, Bose Biswajit, et al. From video sequences to motion panoramas[A]. In: Proceedings of the Workshop on Motion and Video Computing(Motion), Orlando, 2002. 201~207. 被引量:1
  • 5Dufaux Frédéric, Janusz Konrad. Efficient, robust, and fast global motion estimation for video coding[J]. IEEE Transactions on Image Processing, 2000, 9(3): 85~88. 被引量:1
  • 6吴思,张勇东,林守勋,李豪杰.动态场景视频序列中的前景区域自动提取[J].计算机辅助设计与图形学学报,2005,17(2):359-363. 被引量:6
  • 7崔屹编著..图象处理与分析 数学形态学方法及应用[M].北京:科学出版社,2000:171.

二级参考文献9

  • 1Sikora T. The MPEG-4 video standard verification model [J].IEEE Transactions on Circuits and System for Video Technology, 1997, 7(1): 19~31 被引量:1
  • 2Tsaig Y, Averbuch A. Automatic segmentation of moving objects in video sequences: A region labeling approach [J].IEEE Transactions on Circuits and System for Video Technology, 2002, 12(7): 597~612 被引量:1
  • 3Neri A, Colonnese S, Russo G, et al. Automatic moving object and background separation [J]. Signal Processing, 1998, 66(2): 219~232 被引量:1
  • 4Chien S Y, Ma S Y, Chen L G. Efficient moving object segmentation algorithm using background registration technique[J]. IEEE Transactions on Circuits and System for Video Technology, 2002, 12(7): 577~586 被引量:1
  • 5Aach T, Kaup A, Mester R. Statistical model-based change detection in moving video [J] . Signal Processing, 1993, 31(2):165~180 被引量:1
  • 6Habili N, Moini A R. Automatic temporal segmentation for content-based video coding [R]. Adelaide, Australia: Adelaide University, 2000 被引量:1
  • 7Sonka M, Hlavac V, Boyle R. Image Processing, Analysis, and Machine Vision [M]. Florence, KY: Brooks/Cole Publishing,1998 被引量:1
  • 8Thoma R, Bierling M. Motion compensating interpolation considering covered and uncovered background [J]. Signal Processing: Image Communication, 1989, 1: 191 ~ 212 被引量:1
  • 9赵明,李娜,陈纯.采用统计推断的自动视频对象分割[J].计算机辅助设计与图形学学报,2003,15(3):318-323. 被引量:17

共引文献5

同被引文献61

引证文献9

二级引证文献59

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部