首页> 外文期刊>IEEE Transactions on Circuits and Systems for Video Technology >Rectification of Stereoscopic Video for Planar Catadioptric Stereo Systems
【24h】

Rectification of Stereoscopic Video for Planar Catadioptric Stereo Systems

机译:平面折反射立体系统的立体视频校正

获取原文
获取原文并翻译 | 示例

摘要

Planar mirror has been used successfully in the catadioptric stereo systems to capture image pairs with only a single camera. These planar catadioptric stereo (PCS) systems not only provide radiometric advantages over traditional two camera stereo, but also reduce the complexity and cost of acquiring stereoscopic video. One problem with these PCS sensors is that image pairs taken by them are usually not rectified and can not be viewed directly or compressed efficiently. Although several conventional methods have been suggested for rectification, features of the PCS system are not employed to optimize the design and rectified results are not evaluated. In this paper, we investigated features of the PCS system and found that epipolar geometries of certain systems can be approximated by affine. On the basis of this affine property, an effective algorithm is proposed for rectifying image pairs they acquired. This new approach reduces the number of parameters to be estimated from 11 in the conventional method to only four. The greatly reduced number of parameters makes the new method more robust and easier to implement. A PCS system was selected for evaluation and experimental results showed that the proposed method not only outperformed the conventional approaches but also worked quite well both visually and quantitatively for all the image pairs under test
机译:平面反射镜已成功用于折反射立体系统中,仅使用一个摄像机即可捕获图像对。这些平面折反射式立体声(PCS)系统不仅具有比传统的两摄像机立体声系统更强的辐射优势,而且还降低了获取立体视频的复杂性和成本。这些PCS传感器的一个问题是,它们拍摄的图像对通常不会被校正,因此无法直接查看或有效压缩。尽管已建议了几种常规方法进行校正,但并未采用PCS系统的功能来优化设计,并且未评估校正结果。在本文中,我们研究了PCS系统的功能,发现某些系统的对极几何形状可以通过仿射近似。基于这种仿射性质,提出了一种有效的算法来校正其获取的图像对。这种新方法将要估计的参数数量从常规方法中的11个减少到只有4个。参数的数量大大减少,使新方法更强大且更易于实现。选择了PCS系统进行评估,实验结果表明,该方法不仅优于传统方法,而且在所有被测图像对上在视觉和定量上都表现良好

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

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