首页> 外国专利> A depth information-based Stereo/Multi-view Stereo Image Matching Apparatus and Method

A depth information-based Stereo/Multi-view Stereo Image Matching Apparatus and Method

机译:基于深度信息的立体/多视点立体图像匹配装置及方法

摘要

1. This invention belongs to the art set forth in the claims ; The invention depth information based stereo / multiview thing of image matching method and apparatus. ; 2. SUMMARY ; The present invention invention is trying to solve is to decide on one or more pairs of stereo images, the improved accuracy while still variation by determining the variation of the disparity map search range of stereo matching and matching window is adaptively by using the depth information for the reference point obtained by a depth camera in order to improve the accuracy of the disparity map based on the obtained image to minimize the search range stereo / multiview image registration apparatus and method for providing that its purpose is. ; 3. gist ; The invention of the solution of the invention, a depth information-based stereo / multi-viewpoint image matching unit for obtaining a three-dimensional image information to create a multi-view video using the disparity map and camera depth-information obtaining depth information for the multi-viewpoint image and the reference time acquired from the device, Camera calibration means for extracting the time information of each camera using the back-view image; Depth information converting means for converting the depth information in variations of the reference point; And the multi-viewpoint image correction receives the image to match the highest line (epipolar line) of the target image to be matched with the reference point for calculating the disparity map image to the reference point (image rectification) image correction means for performing a also included. ; 4. An important use of the invention ; The present invention is based on the multi-viewpoint video system and scene modeling yiyongdoem.
机译:1.本发明属于权利要求书所述的技术领域。本发明基于深度信息的立体/多视物的图像匹配方法和装置。 ; 2.总结;本发明试图解决的是确定一对或多对立体图像,通过确定立体匹配和视差图的视差图搜索范围的变化来自适应地使用深度信息来确定改进的精度,同时仍保持变化。深度相机为了提高视差图的准确性而基于获取的图像获得的参考点,以最小化搜索范围的立体/多视点图像配准设备和提供其目的的方法。 ; 3.要点本发明的解决方案的发明,基于深度信息的立体/多视点图像匹配单元,用于使用视差图和照相机深度信息获得深度信息来获得三维图像信息以创建多视点视频。从设备获取的多视点图像和参考时间,相机校准装置,用于使用后视图像提取每个相机的时间信息;深度信息转换装置,用于在参考点的变化中转换深度信息;并且,多视点图像校正接收图像以将要匹配的目标图像的最高线(极线)与参考点相匹配,以计算到参考点的视差图图像(图像校正),图像校正装置用于执行也包括在内。 ; 4.本发明的重要用途;本发明基于多视点视频系统和场景建模yiyongdoem。

著录项

  • 公开/公告号KR100776649B1

    专利类型

  • 公开/公告日2007-11-19

    原文格式PDF

  • 申请/专利权人

    申请/专利号KR20050028382

  • 申请日2005-04-06

  • 分类号G06T15/20;

  • 国家 KR

  • 入库时间 2022-08-21 19:54:37

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号