读唇

This method,on one hand extractes the most efficient low features for lipreading,on the other hand,selectes features reasonably to improve their distinguishability.

此方法一方面提取了唇读的最有效的低级语义特征，另一方面更加合理地选择利用了特征的有效可区分性，使得用42维二级视觉特征，对特定人的唇动内容识别正确率达到77.8%。

While sign language and lip reading are very important and often of invaluable help to many people suffering from hearing loss, they aren't always the solution because not everybody knows sign language and some hearing impaired people have not got the skills to lip read.

虽然手语和唇读是非常重要的，往往是宝贵的帮助许多人患有听力丧失，它们并非总是的解决办法，因为不是每个人都知道手语和一些听力受损的人没有得到的技能，以唇读。

The undergraduates who lip-read and heard speech from the same talker were better at identifying the noisy sentences than those who lip-read from one talker and heard speech from another.

结果是，那些对同一位谈话者进行过唇读的学生，在句子识别率上比后者高，即唇读和谈话不是同一人时，对谈话的识别率更低。

It has some specialties as follows:Unsymmetrical lip contour model was used to extract the information of pouting;Chinese pronunciation of vowel and consonant was enhanced to improve the dynamic process;considering the developing of lip-reading techniques,this database was fit for expanding.

不同于现有的数据库，该数据库的设计具有以下特点：采用了非对称唇形轮廓模型，提取了嘴唇突出度的信息；针对汉语音节的特点，增强了汉字音节中信息变化过程；兼顾未来唇读技术的发展，以音节为基本元素，具有可扩充性。

Lip reading ; lip movement ; skin color model ; optical flow ; deformable template.

唇读；唇动；肤色模型；光流；可变模板

Experiments also prove that the features of blocks DCT coefficients in lip region are efficacious to visual single channel lipreading system.

实验还证明了系统中分块的唇区DCT特征对的视觉单通道唇读系统是最有效的。

And a Chinese continuous word lipreading off-line translating system is implemented on the base, it is the largest in scale and highest in recognition power in domestic lipreading system.

并在此基础上完成了一个汉语连续单字唇读识别离线翻译系统，是目前国内最大规模的唇读系统且有最高的识别性能。

A bimodal database HIT Bi-CAVDatabase is established for Chinese lipreading research, and a lipreading recognition off-line system is also finished.

建立了适合进行汉语唇读研究的双模态语料库HIT Bi-CAVDatabase及唇读识别离线系统。

This method,on one hand extractes the most efficient low features for lipreading,on the other hand,selectes features reasonably to improve their distinguishability.With 42-dimensional two-level visual features can get 77.8% rate of lip movement contents recognition for speaker-dependent cases.

此方法一方面提取了唇读的最有效的低级语义特征，另一方面更加合理地选择利用了特征的有效可区分性，使得用42维二级视觉特征，对特定人的唇动内容识别正确率达到77.8%。

To solve the problem of detecting and tracking lip in lipreading, A new real time lip detectionmethod based on the discrepancy of s kin and lip on chrominance and R/G was proposed. First

摘要： 针对图像序列中唇读嘴唇实时检测跟踪问题，提出了一种利用肤色、唇色色度聚类特性，以及R/G 特性差异检测嘴唇的新方法。

I didn't watch TV last night, because it .

昨晚我没有看电视，因为电视机坏了。

Since this year, in a lot of villages of Beijing, TV of elevator liquid crystal was removed.

今年以来，在北京的很多小区里，电梯液晶电视被撤了下来。