针对视频情感识别中存在运算复杂度高的缺点,提出一种基于时空局部二值模式矩(Temporal-Spatial Local Binary Pattern Moment,TSLBPM)的双模态情感识别方法。首先对视频进行预处理获得表情和姿态序列;然后对表情和姿态序列分别提取TSL...针对视频情感识别中存在运算复杂度高的缺点,提出一种基于时空局部二值模式矩(Temporal-Spatial Local Binary Pattern Moment,TSLBPM)的双模态情感识别方法。首先对视频进行预处理获得表情和姿态序列;然后对表情和姿态序列分别提取TSLBPM特征,计算测试序列与已标记的情感训练集特征间的最小欧氏距离,并将其作为独立证据来构造基本概率分配(Basic Probability Assignment,BPA);最后使用Dempster-Shafer证据理论联合规则得到情感识别结果。在双模态表情和姿态情感数据库上的实验结果表明,本文提出的时空局部二值模式矩可以快速提取视频图像的时空特征,能有效识别情感状态。与其他方法的对比实验也验证了本文融合方法的优越性。展开更多
In this paper,we propose temporal Dempster-Shafer theory to handle the combination of uncertainty andtime. In temporal Dempster-Shafer theory,the element of the temporal frame of discernment is defined as an eventthat...In this paper,we propose temporal Dempster-Shafer theory to handle the combination of uncertainty andtime. In temporal Dempster-Shafer theory,the element of the temporal frame of discernment is defined as an eventthat associates a hypothesis with corresponding time interval. And the assignment of belief to subset of the temporalframe of discernment is performed by the mass function. It is a representation and reasoning mechanism that combinesuncertainty and time by the basic frame of Dempster-Shafer theory.展开更多
文摘针对视频情感识别中存在运算复杂度高的缺点,提出一种基于时空局部二值模式矩(Temporal-Spatial Local Binary Pattern Moment,TSLBPM)的双模态情感识别方法。首先对视频进行预处理获得表情和姿态序列;然后对表情和姿态序列分别提取TSLBPM特征,计算测试序列与已标记的情感训练集特征间的最小欧氏距离,并将其作为独立证据来构造基本概率分配(Basic Probability Assignment,BPA);最后使用Dempster-Shafer证据理论联合规则得到情感识别结果。在双模态表情和姿态情感数据库上的实验结果表明,本文提出的时空局部二值模式矩可以快速提取视频图像的时空特征,能有效识别情感状态。与其他方法的对比实验也验证了本文融合方法的优越性。
文摘In this paper,we propose temporal Dempster-Shafer theory to handle the combination of uncertainty andtime. In temporal Dempster-Shafer theory,the element of the temporal frame of discernment is defined as an eventthat associates a hypothesis with corresponding time interval. And the assignment of belief to subset of the temporalframe of discernment is performed by the mass function. It is a representation and reasoning mechanism that combinesuncertainty and time by the basic frame of Dempster-Shafer theory.
