【24h】

Stereo Sequence Coding Based on Segmentation

机译:基于分割的立体声序列编码

获取原文
获取原文并翻译 | 示例

摘要

In this paper, we have discussed various disparity estimation(DE) algorithms. The segmentation-based DE algorithms can provide better performance, and the novel object-based arbitrary shape segmentation is the future developing direction. Meanwhile, a segmentation-based stereo image sequence coding scheme is proposed, which can achieve a higher compression ratio by efficiently exploiting the redundancy within stereo sequences. One image stream is independently coded by using MPEG-type or H.26x-type methods. Only a few reference image frames in the other image stream are coded by adopting the segmentation-based DE algorithms. Most of image frames in the second image stream are not coded and they are reconstructed at the decoder by a frame estimation and interpolation technique.
机译:在本文中,我们讨论了各种视差估计(DE)算法。基于分割的DE算法可以提供更好的性能,新颖的基于对象的任意形状分割是未来的发展方向。同时,提出了一种基于分割的立体声图像序列编码方案,该方案可以通过有效利用立体声序列中的冗余度来实现更高的压缩率。一个图像流通过使用MPEG类型或H.26x类型的方法进行独立编码。通过采用基于分段的DE算法,仅对其他图像流中的少数参考图像帧进行编码。第二图像流中的大多数图像帧不被编码,并且它们在解码器处通过帧估计和内插技术被重建。

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