Shape from polarization images
Shape from polarization images
复制标题
偏振图像的形状
DOI：
10.1016/j.mcm.2004.04.003
复制
发表时间：
1992
期刊：
影响因子：
--
通讯作者：
L. B. Wolff
中科院分区：
文献类型：
--
作者：
L. B. Wolff
我们介绍了一个多视图三维形状重建系统。该系统能够通过独特的神经网络（NN）将少量视图和错误的深度图融合成更完整、更准确的形状表示。该神经网络提供多面体模型的解析映射和学习，以基于多视图深度图近似物体的真实形状。深度图是通过广泛使用的Tsai-Shah形状-从阴影（SFS）算法获得的。它们被认为是要重建的对象的部分3D形状。这项工作的主要观点是，在相对于物体的非固定光源位置下，神经网络使用来自其他视图的深度图信息来最小化一个视图中的深度图误差。从理论上讲，我们将问题表述为由多视图观测形成的深度空间中的非参数（局部）回归。实验上，我们通过分层重构和退火强化得到了精确稳定的结果。我们提供了本文中使用的神经网络的实现。
We introduce a multiple-view 3D-shape-reconstruction system. This system is able to fuse few-view and erroneous depth maps into a more complete and more accurate shape representation using a unique neural network (NN). The NN provides analytic mapping and learning of a polyhedron model to approximate the true shape of an object based on multiple-view depth maps. The depth maps are obtained by a widely used Tsai-Shah shape-from-shading (SFS) algorithm. They are considered as partial 3D shapes of the object to be reconstructed. The main insight of this work is that the NN minimizes the depth map error in one view using depth maps information from other views observed under nonfixed light source positions relative to the object. Theoretically, we formulate our problem as nonparametric (local) regression in depth space formed by multiple view observations. Experimentally, we obtain exact and stable results through hierarchical reconstruction and annealing reinforcement. We provide the implementation of the NN used in this paper at .