权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

Study on Production of Video Images by Computer Controlled Camera Operation and Switching of Plural Camera Images

计算机控制摄像机操作制作视频图像及多摄像机图像切换的研究

基本信息

批准号：
13680456
负责人：
FUKUNAGA Kunio
金额：
$ 2.69万
依托单位：
Osaka Prefecture University
依托单位国家：
日本
项目类别：
Grant-in-Aid for Scientific Research (C)
财政年份：
2001
资助国家：
日本
起止时间：
2001 至 2002
项目状态：
已结题

来源：
https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-13680456/
关键词：
Computer Vision Production of Video Imags Distant Learning System Recognition of Human Motion Video Image Processing

项目摘要

High-speed digital data transmission becomes possible accompanied with the development of network techniques, and distant learning systems are developed. In this project, we propose a method of producing vivid video images of the distance learning using a computer controlled camera operation and switching of plural camera images.(1) Shooting the scene based on distribution of spatiotemporal mutual informationWe define spatiotemporal mutual information on the pixels of a given video image on the basis of information theory, which can be interpreted as the theoretical estimation of interested spots for human being. Using the distribution of the spatiotemporal mutual information, we propose an algorithm of computer controlled camera operation and switching of plural camera images to take vivid video images of distance learning with real time image processing.(2) Shooting the scene based on situation understanding and evaluation of video imagesA shooting camera system estimates actions of a teacher in front of the blackboard considering the movement of the head and the hands and written characters on a blackboard. Each camera chooses a shooting area based on the actions of the teacher, then computer controlled switching system selects the best image among plural imagss by evaluation each image.

随着网络技术的发展，高速数字数据传输成为可能，远程学习系统也随之发展起来。在这个项目中，我们提出了一种使用计算机控制的摄像机操作和多个摄像机图像的切换来产生生动的远程学习视频图像的方法。(1)基于时空互信息分布的场景拍摄我们在信息论的基础上定义了给定视频图像像素的时空互信息，它可以解释为对人类感兴趣点的理论估计。利用时空互信息的分布特性，提出了一种计算机控制摄像机操作和切换多个摄像机图像的算法，以获得逼真的远程教学视频图像，并进行真实的时间图像处理。(2)基于视频图像的情况理解和评估的场景拍摄拍摄摄像机系统考虑到头部和手部的运动以及黑板上的书写字符来估计黑板前的教师的动作。每个摄像机根据教师的动作选择一个拍摄区域，然后计算机控制的切换系统通过评估每个图像从多个图像中选择最佳图像。