3D structure from 2D motion

被引:112
作者
Jebara, T [1 ]
Azarbayejani, A [1 ]
Pentland, A [1 ]
机构
[1] MIT, Media Lib, Cambridge, MA 02139 USA
关键词
D O I
10.1109/79.768574
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
The structure from motion (SfM) approach is presented for recursive estimation of motion, structure, and camera geometry in a nonlinear dynamic system framework. These are used to assess the accuracy and stability of the technique. Some practical and real time applications that have been encountered are discussed and the reliability and flexibility of the approach in these settings are analyzed. Results from an independent evaluation study conducted by industry where the proposed SfM algorithm compared favorably to alternative approaches are also presented.
引用
收藏
页码:66 / 84
页数:19
相关论文
共 63 条
[1]  
[Anonymous], 1974, APPL OPTIMAL ESTIMAT
[2]  
[Anonymous], 1993, SPRINGER SERIES INFO
[3]  
[Anonymous], 1993, Three-Dimensional Computer Vision: A Geometric Viewpoint
[4]  
[Anonymous], MACH VISION APPL
[5]   MAINTAINING REPRESENTATIONS OF THE ENVIRONMENT OF A MOBILE ROBOT [J].
AYACHE, N ;
FAUGERAS, OD .
IEEE TRANSACTIONS ON ROBOTICS AND AUTOMATION, 1989, 5 (06) :804-819
[6]   VISUALLY CONTROLLED GRAPHICS [J].
AZARBAYEJANI, A ;
STARNER, T ;
HOROWITZ, B ;
PENTLAND, A .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1993, 15 (06) :602-605
[7]   RECURSIVE ESTIMATION OF MOTION, STRUCTURE, AND FOCAL LENGTH [J].
AZARBAYEJANI, A ;
PENTLAND, AP .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1995, 17 (06) :562-575
[8]  
AZARBAYEJANI A, 1996, VIS P SIGGR 96 NEW Y, P155
[9]  
AZARBAYEJANI AJ, 1994, 2 CAD BAS VIS WORKSH
[10]  
AZARBAYEJANI AJ, 1997, THESIS MIT