茶峒:VisionLab-上海交通大学计算机视觉实验室

刘允才教授，1990年在美国伊利诺伊大学（UIUC）电气计算机工程系获得博士学位。1990-1991年在美国伊利诺斯大学贝克曼科学研究院从事博士后研究。1991年加入日本住友电气工业株式会社，任研究员系统顾问、研究员总顾问。2000年10月加盟上海交通大学，被聘为国家教育部长江学者奖励计划特聘教授，博士生导师，国际学术期刊《Pattern Recognition》编委，中国图像与图形学学会学术委员会委员、学会理事。刘允才教授长期从事于计算机视觉理论方面的研究工作，在采用二维视频图像的刚体三维空间运动估值方面做出过原创性的研究。他导出了采用图像线元求解空间刚体三维运动参数的必要条件和完整的算法（非线性算法）；在此基础上，又首次导出相应的线性算法，为三维空间运动估值方面的研究奠定了理论基础。刘允才教授还首次提出了基于二维——三维线对应元的图像载体空间定位方法，该理论被有关领域的学者广泛引用。在采用图像角元的刚体三维运动估值方面，他原创性地提出了由单幅图像角元直接求解空间物体结构的算法，并应用于刚体三维空间运动估值。他还第一个提出并证明常参数运动是一个柱面螺旋运动，并将这一结论应用于长序列图像的三维空间运动估值。在近期研究的连接刚体运动分析方面，他导出了在不存在运动约束的情况下，连接刚体初始姿态定位、连接点位置估计和序列图像的连接刚体运动分析方法；此外，他在图像特征提取、图像特征匹配、摄像机与系统定标等方面进行了大量的研究。目前，正致力于弹性连接体和人体运动分析方面的研究。刘允才教授的研究兴趣包括非刚体物体的三维运动估值、视觉导航与空间定位、三维重建、异构图像配准、系统定标、计算机辅助外科手术、数字地图自动生成、智能交通等，承担了国家自然基金、国家973课题、国家863、国家教育部、上海市科委、国际合作等项目；联合出版著作3部，在包括IEEE Trans. PAMI、IEEE Trans. Image Processing、IEEE Trans. Intelligent Transportation Systems、Pattern Recognition、Physic Review、CVPR、ICCV、ICPR等学术期刊与会议上发表论文200余篇；申请发明专利65项，授权35项；获得2002年度教育部提名国家科学技术奖自然科学二等奖，上海市2003、2004、2005、2009年度科学进步二等奖；被评为上海市优秀博士论文奖导师、全国优秀博士论文提名奖导师。
办公地点:上海交通大学闵行校区电信大楼二号楼431室
电话:(86-21)3420-4340
传真:(86-21)3420-4340
实验室地点:上海交通大学闵行校区电信大楼二号楼233室
电话:(86-21)3420-4028
代表性论文
[1] Xiaoyun Zhang, Yuncai Liu and Thomas S. Huang, "Motion Analysis of Articulated Objects from Monocular     Images", IEEE trans. Pattern Anal. Mach. Intell., Vol 28, No.4, 2006.
[2] Yuncai Liu and Thomas S. Huang, "Vehicle-Type Motion Estimation from Multi-Frame Images," IEEE trans.    Pattern Anal. Mach. Intell., Vol 15, No. 8, Aug., 1993.
[3] Yuncai Liu, Thomas S. Huang and O. D. Faugeras, "Determination of Camera Location from 2D to 3D Line and     Point Correspondences," IEEE trans. Pattern Anal. Mach. Intell., Vol 12, No. 1, Jan. 1990.
[4] Zijian Zhao, Yuncai Liu, Zhengyou Zhang, "Camera Calibration with 3 Non-collinear Points Under Special     Motions", IEEE Trans. on Image Processing, vol.17, no.12, pp.2393-2402, 2008.
[5] Shuhan Shen, Wenhuan Shi, Yuncai Liu, "Monocular 3D Tracking of Inextensible Deformable Surfaces under     L2-Norm," IEEE Trans. on Image Processing, Accepted.
[6] Shuhan Shen, Wenjuan Ma, Wenhuan Shi, Yuncai Liu, "Convex Optimization for Non-Rigid Stereo     Reconstruction," IEEE Trans Image Processing, Accepted.
[7] Xu Zhao,Yun Fu, Yuncai Liu, Thomas S. Huang, "Human Pose Estimation with Regression by Fusing MultiView     Visual Information," IEEE Trans. on Circuits and Systems for Video Technology, Accepted.
[8] Xu Zhao, Yun Fu, Huazhong Ning, Yuncai Liu, and Thomas S. Huang, "Text from Corners: A Novel Approach     to Detect Text and Caption in Videos," IEEE Trans. on Image Processing, Accepted.
[9] Xu Zhao, Yuncai Liu, "Temporal-Spatial Combined Local Experts for Sparse Gaussian Process Regression",    IEEE Trans. on Image Processing, Accepted.
[10] Qingjie Kong, Zhipeng Li, Yikai Chen and Yuncai Liu, "An approach to urban traffic state estimation by      fusing multisou-rce information,"IEEE Trans. on Intelligent Transportation Systems, vol. 10,      no. 3, Sept. 2009.
[11] Yuncai Liu, "An Automation System: Generation of Digital Map Data from Pictorial Resources," Pattern     Recognition, Vol.35, No.9, 2002.
[12] Hailang Pan, Yuncai Liu，"Motion Estimation of Elastic Articulated Objects from Points and Contours      with Volume Invariable Constraint," Pattern Recognition, vol. 41, no. 2, pp. 458-467, 2008.
[13] Shuhan Shen, Yuncai Liu, " Monocular 3D Tracking of Deformable Surfaces Using Sequential Second Order      Cone Programming," Pattern Recognition, vol. 43, no. 1, pp. 244-254, 2010.
[14] Xu Zhao, Yuncai Liu, "Generative tracking of 3D human motion by hierarchical annealed genetic algorit-
hm,"Pattern Recognition, vol. 41, no. 8, pp. 2470,2008.
[15] Fanhuai Shi, Jianhua Wang, Jing Zhang, Yuncai Liu, "Motion Segmentation of Multiple Translating Objec-
ts from Line Correspondences" , Pattern Recognition, vol.38, no.10, pp.1775-1778, 2005.
[16] Jianhua Wang, Fanhuai Shi, Jing Zhang and Yuncai Liu , "A new calibration model of camera lensdistort-
ion," Pattern Recognition, vol. 41, no. 2, pp.607-615, 2008.
[17] Yuncai Liu and Thomas S. Huang, "Three-Dimensional Motion Determination from Real Scene Image Using      Straight Line Correspondences," Pattern Recognition, Vol 25, No 6, June 1992.
[18] Yuncai Liu and Thomas. S. Huang, "Determine Straight Line Correspondences from Intensity Images," Pat-
tern Recognition,Vol 24, No. 6, 1991.
[19] Yuncai Liu, Xiaoyun Zhang and T. S. Huang, "Estimation of 3D Structure and Motion fromImage Corners,"-     Pattern Recognition, 36 (6): 1269-1277, 2003.
[20] Zhi peng Li, Yuncai Liu, "Analysis of the stability and density waves of traffic flow model in an ITS      environment," European Physical Journal B, vol.53, pp.367-374, 2006.
 
链接:承担的科研项目与获奖情况

视觉系统标定
Calibration of Vision System
摄像机标定 Camera calibration
手眼标定 Hand-eye calibration
工具标定 Instrument calibration
手眼式机器人辅助全膝关节置换手术系统
Hand-Eye Robotic System for TKR
全膝关节置换手术是一种关节功能的重建手术。其关键操作是对病人的股骨做五次定向切割与胫骨横向切割，并用合适的假体替换受损关节。
城市路网交通状态估计分析系统
Analytic System of Traffic Status Estimation
基于SCATS数据的交通状态估计 Estimation from SCTAT Data
基于出租车数据的交通状态估计 Estimation from Taxi Data
虚拟环交通检测 Traffic Detection by Virtual Loop
弹性连接刚体三维建模和运动估计
3D Modeling and Motion of Elastic Articulated Objects
人体三维建模 3D Human Body Modeling
基于立体视觉的人肢体建模 Human Arm Modeling From Stereo Images
虚拟人模型制作 Virtual Human Model
连接刚体与人体运动分析

连接刚体运动分析
Articulated rigid object motion estimation
利用单目图像进行人体运动分析
Human motion analysis with monocular image
交通信息实时视频检测系统
Real-time traffic information video detection system
交通信息实时视频检测 RT detection of traffic infomation
硬件实现:DSP
城市路网交通监控系统
Municipal Traffic Information System
地图匹配 Map-Match
交通流模型 Traffic Model
拥堵状态估计 Congestion Estimation
人脸建模与颜色变换
Face Modeling and Color Transfering
由正交图像建立三维人脸模型
3D face modeling from orthogonal images
图像之间的颜色变换
Color transfer between images
Last Updated on 07/06/2010                                                                      copyright © VisionLab 2006                                       沪交ICP备2010877
2011年，刘允才教授继续招收博士研究生，欢迎报考！报考时请暂时填报杨根科教授，录取后自动转入刘允才教授名下！