Depth Restoration From RGB-D Data via Joint Adaptive Regularization and Thresholding on Manifolds
Depth Restoration From RGB-D Data via Joint Adaptive Regularization and Thresholding on Manifolds
复制标题
通过联合自适应正则化和流形阈值从 RGB-D 数据恢复深度
DOI：
10.1109/tip.2018.2872175
复制
发表时间：
2019-03
影响因子：
10.6
通讯作者：
Gao Wen
中科院分区：
文献类型：
--
作者：
Liu Xianming;Zhai Deming;Chen Rong;Ji Xiangyang;Zhao Debin;Gao Wen
本文结合局部流形和非局部流形的特点，提出了一种新的基于RGB-D数据的深度恢复算法，提供了深度图局部和非局部几何的低维参数化。具体地说，一方面定义了局部流形模型以利于像素在深度上的局部邻域关系，并根据该模型引入流形正则化来促进沿流形结构的平滑。另一方面，基于面片的流形的非局部特性可以用来构建高度数据自适应的正交基，以提取细长的图像模式，考虑到流形中的自相似结构。我们进一步定义了3D自适应正交谱基中的流形阈值算子--局部流形和非局部流形的离散拉普拉斯特征向量--以仅保留用于深度图恢复的低图形频率。最后，我们提出了一种统一的交替方向乘子法优化框架，该框架巧妙地将自适应流形正则化和阈值联合应用于深度图恢复反问题的正则化。实验结果表明，该方法在客观质量评价和主观质量评价方面均优于目前最先进的评价方法。
In this paper, we propose a novel depth restoration algorithm from RGB-D data through combining characteristics of local and non-local manifolds, which provide low-dimensional parameterizations of the local and non-local geometry of depth maps. Specifically, on the one hand, a local manifold model is defined to favor local neighboring relationship of pixels in depth, according to which, manifold regularization is introduced to promote smoothing along the manifold structure. On the other hand, the non-local characteristics of the patch-based manifold can be used to build highly data-adaptive orthogonal bases to extract elongated image patterns, accounting for self-similar structures in the manifold. We further define a manifold thresholding operator in 3D adaptive orthogonal spectral bases—eigenvectors of the discrete Laplacian of local and non-local manifolds—to retain only low graph frequencies for depth maps restoration. Finally, we propose a unified alternating direction method of multipliers optimization framework, which elegantly casts the adaptive manifold regularization and thresholding jointly to regularize the inverse problem of depth maps recovery. Experimental results demonstrate that our method achieves superior performance compared with the state-of-the-art works with respect to both objective and subjective quality evaluations.
登录
查看更多内容
通过联合颜色引导的内部和外部正则化实现深度超分辨率
影响因子：
10.6
作者：
Liu Xianming;Zhai Deming;Chen Rong;Ji Xiangyang;Zhao Debin;Gao Wen
通讯作者：
Gao Wen
影响因子：
7.3
作者：
Deming Zhai;Xianming Liu;Xiangyang Ji;Debin Zhao;S. Satoh;Wen Gao
通讯作者：
Deming Zhai;Xianming Liu;Xiangyang Ji;Debin Zhao;S. Satoh;Wen Gao
参数化局部多视图汉明距离度量学习
影响因子：
8
作者：
Deming Zhai;Xianming Liu;Hong Chang;Yi Zhen;Xilin Chen;Maozu Guo;Wen Gao
通讯作者：
Wen Gao
DOI：
10.1016/j.cviu.2008.09.003
发表时间：
2009-02
期刊：
Comput. Vis. Image Underst.
影响因子：
--
作者：
G. Peyré
通讯作者：
G. Peyré
影响因子：
10.6
作者：
Xianming Liu;Debin Zhao;Ruiqin Xiong;Siwei Ma;Wen Gao
通讯作者：
Xianming Liu;Debin Zhao;Ruiqin Xiong;Siwei Ma;Wen Gao