Research Content/Issue
An accurate three dimensional sign language data are collected
by the data capturing method based on motion tracking with two data
gloves and motion trackers. This method helped us to accomplish
a total 5596 Chinese Sign Language vocabularies and built up a visual
stereo database of Chinese Sign Language.
Chinese Sign language method based on virtual human synthesis technology.
In the virtual human synthesis technology, we proposed a face synthesis
method that takes into account all of the facial characteristics
through virtual human synthesis technology and a facing Mpeg4 optional
facial grid drive method. Sign language editing methods using reference
points, machine learning method for speech and lip motion synchronization,
and data mining method that finds and synchronizes speech with lip
movement sign language method are respectively presented. The above
methods helped us to obtain the data on virtual reality for visualizing
human sign language on a facing motion-monitoring device so the
virtual human can realize sign language expressions by synchronizing
with speech, lip movement and gestures.
Research Achievement
 Technology
patents
| Patent Name |
Kind |
Application Number |
Proposer |
Time |
Affiliation
organization |
Current Status |
| The sign language vocabulary editing method based on a control
point of motion |
Invention |
01115947.2 |
2001/6/6 |
Gao Wen, Wang Zhaoqi |
Institute of Computing Technology |
First (trial) approval |
| A speech driving the cartoon-like face method based on combined
statistics and rules |
Invention |
02140286.8 |
2002/7/3 |
Gao Wen, Wang Zhaoqi, Yiqiang Chen |
Institute of Computing Technology |
First (trial) approval |
| The cartoon-like face modeling method with any formats based
on the MPEG-4 |
Invention |
02140285.x |
2002/7/3 |
Gao Wen, Wang Zhaoqi, Dalong Jiang |
Institute of Computing Technology |
First (trial) approval |
| The sign language translating method with a medium language
agency |
Invention |
02121369.0 |
2002/6/17 |
Gao Wen, Jiyong Ma, Chunli Wang, Jiangqin Wu,
Xilin Chen |
Institute of Computing Technology |
First (trial) approval |
Reward Situation
In Nov. of 2002, it won second place for the Science and Technology
Progress in Beijing based on the Chinese sign language recognition
and synthesis research of multimodal perception theory, and it was
also recommended by the Beijing Science Committee to apply for award
of State Science and Technology Progress.
Publications
[1] 王兆其，高文，基于虚拟人合成技术的中国手语合成方法，《软件学报》，13(10):2051-2056,2002。
[2] 王兆其，杨长水，高文，基于 VRML 的中国手语三维显示，《计算机研究与发展》，2001， 已录用。
[3] 王兆其，高文，陈益强，杨长水，左力，姜大龙，李治国，虚拟人行为交互方法的研究，第一届全国虚拟现实与可视化学术会议CCVRV'01,2001.11,北京。
[4] 陈益强，高文，王兆其，杨长水，姜大龙，"多模式语音合成"，第六届全国人机语音通讯学术会议NCMMSC6，p163-168，2001.11，深圳
[5] 陈益强，高文，王兆其，姜大龙，基于机器学习的语音驱动人脸动画方法，《软件学报》，2001，已录用。
[6] 陈益强，高文，王兆其，左力，姜大龙，基于数据挖掘的语音驱动三维人脸动画合成 ，系统仿真学报，Vol.14, No.4 496-500，第一届全国虚拟现实及可视化技术会议优秀论文
[7] 杨长水，王兆其，高文，基于WEB的手语新闻虚拟主持人的研究与实现，第一届全国虚拟现实与可视化学术会议CCVRV'01,2001.11,北京。
[8] 姜大龙，王兆其，高文，基于MPEG-4的三维人脸动画实现方法,第一届全国虚拟现实与可视化学术会议CCVRV'01,2001.11,北京。
[9] 王兆其，高文，徐燕，一种基于传感器的人体上肢运动实时跟踪方法，《计算机学报》，24(6):616-619，2001。
[10] 高文，陈熙霖，马继勇，王兆其，基于多模式接口技术的聋人与正常人交流系统，《计算机学报》，23(12):1253- 1260,
2000.
[11] 王兆其，虚拟人合成研究综述，《中国科学院研究生院学报》，2000, 17(2):89-98.
[12] 王兆其，杨长水，虚拟主持人的设计与制作，动画与特技，2002，已录用
[13] 李治国，姜大龙，高文，王巍，全自动的基于MPEG-4的任意拓扑结构人脸动画实现方法，计算机辅助设计与图形学学报
[14] Yiqiang Chen, Wen Gao, Zhaoqi Wang, Li Zuo, Speech driven MPEG-4
based face animation via neural network(286KB), The Second IEEE Pacific-Rim
Conference On Multimedia 2001, October 24-26, 2001, Beijing, China.
[15] Yiqiang Chen，Wen Gao，TingShao Zhu，Charles Ling，Learning prosodic
patterns for mandarin text to speech，The Journal of Intelligent Information
System，Vol 19, No 1 95-109，Jun-02
[16] Yiqiang Chen，Wen Gao，Zhaoqi Wang，Changshui Yang，Dalong Jiang
Text to Aatar in Multimodal Human Computer Interface，Asia-Pacific
Human Computer Interface (APCHI2002)，2002，accepted
[17] Yiqiang Chen，Wen Gao，Zhaoqi Wang，Li Zuo，Speech Driven Mpeg4 based
Face Animation via Neural Network，In Proc. IEEE Pacific Conference
on Multimedia, (PCM2001)，pp1108-1113，2001/10/24-26
[18] Yiqiang Chen，Wen Gao，Zhaoqi wang，Jun Miao，Dalong Jiang，Mining
Audio/Visual Database for Speech Driven Face Animation，In Proc. IEEE
International Conference on System Man and Cybernetics (SMC2001)，pp2638-2643，Oct-01
[19] Yiqiang Chen，Wen Gao，TingShao Zhu，Jiyong Ma，Multi-strategy Data
mining framework for mandarin prosodic pattern International Conference
on Spoken Language Processing (ICSLP2000)，Vol.2, 59-62，2002
[20] Yiqiang Chen，Wen Gao，Hand Gesture Recognition Based on Decision
Tree，International Symposium of Chinese on Spoken Language Processing
(ISCSLP2000)，pp209-303，2000
[21] Yiqiang Chen，Wen Gao，TingShao Zhu，Linguistic Feature selection
for mandarin fundament frequency，International Symposium of Chinese
on Spoken Language Processing (ISCSLP2000)，pp137-141，2000
[22] Yiqiang Chen，Wen Gao，Data mining in Speech Synthesis In Proceeding
of SCI2000 Vol.III, pp58-63,Orlando, Florida, USA，2000/7/23-24
[23] Dalong Jiang，Wen Gao，Zhiguo Li，Zhaoqi Wang，Animating Arbitrary
Topology 3D Facial Model Using the MPEG-4 FaceDefTables The fourth
International Conference on Multi-modal Interface, IEEE ICMI'2002，Pittsburgh,
USA，2002-10-14
[24] Dalong Jiang, Zhiguo Li, Zhaoqi Wang, Wen Gao, Animating 3D Facial
Models with the Mpeg-4 FaceDefTables, 35th Annual Simulation Symposium,
2001.
[25] Xu Lin,Gao Wen,Yao Hongxun,Ma Jiyong,Study on Machine Translation
of Rule-Based Chinese Sign Language,Journal of Computer Science and
Technology,2000
[26] Xu Lin,Gao Wen,Human-Computer Chinese Sign Language Interaction
System International Journal of Virtual Reality,2000
[27] 徐琳,高文,面向机器翻译的中国手语的理解与合成,计算机学报,Vol 23,No.1,pp.60-65,2000 |
