Insights into machine lip reading

Insights into machine lip reading
复制标题

机器唇读的见解

DOI:
10.1109/icassp.2012.6288999
复制
发表时间:
2012
期刊:
2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
影响因子:
--
通讯作者:
B. Theobald
B. Theobald
中科院分区:
--
文献类型:
--
作者:
Yuxuan Lan;R. Harvey;B. Theobald

文献摘要

参考文献

被引文献

相似文献

计算机唇读是信号处理面临的巨大挑战之一。信号不仅有噪声,而且是可变的。然而,与人类唇读器的性能进行比较几乎是未知的。部分原因是人类唇读器的缺乏,部分原因是大多数自动系统只处理琐碎的数据,因此不能代表人类语音。在这里,我们使用连接的单词生成一个多视图数据集,该数据集可以由基于线性预测跟踪器和主动外观模型以及人类唇读器的自动系统进行分析。我们设计的自动系统的视位准确度约为 46%,与较差的专业人类唇读器相当。然而,与人类唇读器不同的是,我们的系统擅长猜测其错误性。
Computer lip-reading is one of the great signal processing challenges. Not only is the signal noisy, it is variable. However it is almost unknown to compare the performance with human lip-readers. Partly this is because of the paucity of human lip-readers and partly because most automatic systems only handle data that are trivial and therefore not representative of human speech. Here we generate a multiview dataset using connected words that can be analysed by an automatic system, based on linear predictive trackers and active appearance models, and human lip-readers. The automatic system we devise has a viseme accuracy of ≈ 46% which is comparable to poor professional human lip-readers. However, unlike human lip-readers our system is good at guessing its fallibility.
DOI: --
发表时间: 2008-09
期刊: --
影响因子: --
作者:
S. Cox;R. Harvey;Yuxuan Lan;Jacob L. Newman;B. Theobald
通讯作者: S. Cox;R. Harvey;Yuxuan Lan;Jacob L. Newman;B. Theobald
比较唇读的视觉特征
DOI: --
发表时间: 2009
期刊: AVSP-2009
影响因子: --
作者:
Lan Y
通讯作者: Lan Y