Variational phasor mean field model for object recognition
Variational phasor mean field model for object recognition
复制标题
DOI：
10.1109/isccsp.2008.4537275
复制
发表时间：
2008-03
期刊：
影响因子：
--
通讯作者：
Haruhisa Takahashi
中科院分区：
文献类型：
--
作者：
Haruhisa Takahashi
马尔可夫随机场的变分相量平均场模型（VPMF）能很好地反映站点间的相关性和边缘分布。该网络是由复杂的方程，其中包括相位方程和变分的平均场方程，因此，VPMF不仅能够提高平均场近似的精度，但也给额外的相关关系与余弦的相位差的单元之间。在这份报告中，我们讨论VPMF作为一个对象识别工具，并通过计算机实验表明，它提供了有效的学习方法。
The variational phasor mean field model (VPMF) for Markov random fields can well represent marginal distribution as well as correlation among the sites. The network is represented by complex equations, which consist of phase equations and variational mean-field equations; thus the VPMF enables not only to improve the accuracy of the mean field approximation but also to give additional correlational relation between units with the cosine of the phase differences. In this report we discuss VPMF as an object recognition tool, and show that it provides efficient learning methods through computer experiments.