人脸识别研究进展

人脸识别技术在各种应用的推动下发展很快。而近几年中，为避免各种实际应用环境的限制，非约束环境下的人脸识别也得到了很大的发展。本文概述了人脸识别的一些主要方法，讨论了影响人脸识别的几个主要因素及其处理方法以及基于视频的人脸识别技术，最后指出了鲁棒的人脸识别技术的研究方向和发展前景。     

眼球的自动定位

提出了一种由粗到细自动定位人脸的一个重要特征即眼球圆心的方法，该方法综合运用局部二值化，霍夫变换，像纱聚类，边缘提取等技术，对均匀或非均匀光照下拍摄或戴眼镜的照片都能获得很高的定位精度，适用于处理证件照。     

Counting unrooted maps on the plane

A planar map is a 2-cell embedding of a connected planar graph, loops and parallel edges allowed, on the sphere. A plane map is a planar map with a distinguished outside (“infinite”) face. An unrooted map is an equivalence class of maps under orientation-preserving homeomorphism, and a rooted map is a map with a distinguished oriented edge. Previously we obtained formulae for the number of unrooted planar n-edge maps of various classes, including all maps, non-separable maps, eulerian maps and loopless maps. In this article, using the same technique we obtain closed formulae for counting unrooted plane maps of all these classes and their duals. The corresponding formulae for rooted maps are known to be all sum-free; the formulae that we obtain for unrooted maps contain only a sum over the divisors of n. We count also unrooted two-vertex plane maps.     

Error Concealment Based on Semantic Prioritization with Hardware‐Based Face Tracking

With video compression standards such as MPEG‐4, a transmission error happens in a video‐packet basis, rather than in a macroblock basis. In this context, we propose a semantic error prioritization method that determines the size of a video packet based on the importance of its contents. A video packet length is made to be short for an important area such as a facial area in order to reduce the possibility of error accumulation. To facilitate the semantic error prioritization, an efficient hardware algorithm for face tracking is proposed. The increase of hardware complexity is minimal because a motion estimation engine is efficiently re‐used for face tracking. Experimental results demonstrate that the facial area is well protected with the proposed scheme.     

基于身份感知模型的素描人脸识别方法

针对素描图像和光学图像之间存在较大的模态差异这一问题,提出了一种基于身份感知模型的素描人脸识别方法,实现跨模态图像生成和素描人脸识别。该方法应用新的感知损失来监督图像生成网络,生成更好的跨模态图像,减少模态差异带来的识别精度损失,并通过三元组损失来正则化类内和类间距离,增强识别模型的性能,用联合训练策略提升素描人脸识别能力。在UoM-SGFSv2、e-PRIP等素描人脸数据集上的实验结果表明,该方法识别效果优于其他对比算法。     

人脸识别技术综述

文章首先对人脸识别技术进行了介绍，其次回顾了人脸识别研究的发展历程及识别方法的基本分类，然后对当前主流的人脸识别方法展开了详细的论述，最后提出了人脸识别技术面临的问题及研究方向。     

An improved exponential transformation for nearly singular boundary element integrals in elasticity problems

This paper presents an improved exponential transformation for nearly singular boundary element integrals in elasticity problems. The new transformation is less sensitive to the position of the projection point compared with the original transformation. In our work, the conventional distance function is modified into a new form in the polar coordinate system. Based on the refined distance function, an improved exponential transformation is proposed in the polar coordinate system. Moreover, to perform integrations on irregular elements, an adaptive integration scheme considering both the element shape and the projection point associated with the improved transformation is proposed. Furthermore, when the projection point is located outside the integration element, another nearest point is introduced to subdivide the integration elements into triangular or quadrilateral patches of fine shapes. Numerical examples are presented to verify the proposed method. Results demonstrate the accuracy and efficiency of our method.     

网络环境下高职院校学生思想政治教育的创新刍议

本文将明确网络环境下高职院校学生思想政治教育进行创新的重要性和必要性，探究网络环境下高职院校学生思想政治教育面临的挑战，并提出一些在网络环境下高职院校学生思想政治教育进行创新的途径。