Original document(17 pages)  中文版
    The invention discloses a video content expressing method based on the significant element of space time, which is based on the mechanism of biological visual attention, and maps the former characteristic space into the illegible information granule space by the method of illegible information granulating, to realize the distinct analysis of visual information effectively and pick up the distinct region conforming to the vision features as the analysis body for time consistency of the subunit. On the base, the combination of sub units is conduct under the different impacts of distinction and background regions on the content description, to get the distinct space unit with consistent content. Experimental results show the effectiveness and semantic nature of the mode that distinct space unit pick up by the invention is taken as the video content.
Application Number
申请号
200610113909 Application Date
申请日
2006.10.20
Title 名称 Video frequency content expressing method based on space-time remarkable unit
Publication Number
公开号
1945628 Publication Date
公开日
2007.04.11
Approval Pub. Date Granted Pub. Date
International Classification 分类号 G06T7/00;G06T5/00
Applicant(s) Name
申请人
Beijing Jiaotong Univ.
Address 地址
Inventor(s) Name 发明人 Xu De;Liang Congyan;Yang Xu
Attorney & Agent 代理人 wuke yu maoyan sheng
More information 更  多  信  息


 Related patents information
Method for calculating image object size constance
Google
Note:All patent data come from State Intellectual Property Office of the People's Republic of China. If there were discrepancies between here and the State Intellectual Property office, the later is more accurate. The patent data is only for public exchange and learning purposes. We are not responsible for the adverse consequences with unverified use of the data.