Unsupervised adaptive sign language recognition based on hypothesis comparison guided cross validation and linguistic prior filtering

Unsupervised adaptive sign language recognition based on hypothesis comparison guided cross validation and linguistic prior filtering
复制标题

基于假设比较引导交叉验证和语言先验过滤的无监督自适应手语识别

DOI:
10.1016/j.neucom.2014.08.032
复制
发表时间:
2015-02
期刊:
影响因子:
6
通讯作者:
Yipeng Wang
Yipeng Wang
中科院分区:
计算机科学2区
文献类型:
--
作者:
Xiaokang Yang;Yongzheng Zhang;Xiang Xu;Yipeng Wang

文献摘要

参考文献

相似文献

手语自适应是手语识别系统的重要组成部分,因为一个固定的系统不可能对所有类型的手语都有很好的识别效果。在有监督的签名者自适应中,必须显式地收集标记的自适应数据。为了跳过签名者自适应中的数据收集过程,提出了一种新的无监督自适应方法,即假设比较指导的交叉验证方法。该方法不仅解决了待标记数据集与自适应数据集重叠的问题,而且采用了一个附加的假设比较步骤来降低自适应数据集的噪声率。我们还利用语言的先验知识下采样的适应数据列表,以进一步降低噪声率。为了评估所提出的方法的有效性,CASIIE-SL数据库的形成,这是第一个专门的数据集,无监督签名适应我们所知道的。实验结果表明,与自学习方法和交叉验证方法相比,该方法的相对误词率分别降低了3.93%和4.05%。该方法虽然是针对签名者自适应提出的,但也可以直接应用于说话人自适应和书写人自适应。
Signer adaptation is important for sign language recognition systems because a fixed system cannot perform well on all kinds of signers. In supervised signer adaptation, the labeled adaptation data must be collected explicitly. To skip the data collecting process in signer adaptation, we propose a novel unsupervised adaptation method, namely the hypothesis comparison guided cross validation method. The method not only addresses the problem of the overlap between the data set to be labeled and the data set for adaptation, but also employs an additional hypothesis comparison step to decrease the noise rate of the adaptation data set. We also utilize linguistic prior knowledge to down sample the adaptation data list to further decrease the noise rate. To evaluate the effectiveness of the proposed method, the CASIIE-SL-Database is formed, which is the first specialized data set for unsupervised signer adaptation to the best of our knowledge. Experimental results show that the proposed method can achieve relative word error rate reductions of 3.93% and 4.05% respectively compared with self-teaching method and cross validation method. Though the method is proposed for signer adaptation, it can also be applied to speaker adaptation and writer adaptation directly.
DOI: 10.1109/tkde.2005.186
发表时间: 2005-11
影响因子: 8.9
作者:
Zhi-Hua Zhou;Ming Li
通讯作者: Zhi-Hua Zhou;Ming Li
DOI: 10.1007/978-1-349-22599-6_7
发表时间: 2015-10
期刊: --
影响因子: --
作者:
Douglas O'Shaughnessy
通讯作者: Douglas O'Shaughnessy
DOI: 10.1007/11492429_63
发表时间: 2005-06
期刊: --
影响因子: --
作者:
Jörg Zieren;K. Kraiss
通讯作者: Jörg Zieren;K. Kraiss
DOI: 10.1007/978-81-322-3972-7_20
发表时间: 2020
期刊: Fundamentals of Artificial Intelligence
影响因子: --
作者:
Prof. K. R. Chowdhary
通讯作者: Prof. K. R. Chowdhary
DOI: 10.1007/978-1-4471-0285-4
发表时间: 2012-10
期刊: --
影响因子: --
作者:
Shigeo Abe DrEng
通讯作者: Shigeo Abe DrEng