RI | Research | Projects | MBV

Text only version of this site

Modeling by Videotape (MBV) Head: Takeo Kanade Contact: Takeo Kanade (tk@cs.cmu.edu) Mailing address: Carnegie Mellon University Robotics Institute 5000 Forbes Avenue Pittsburgh, PA 15213

This page last updated - January 1999.

Project Description

Imagine that you give me a videotape of your room that you have made by walking around with your hand-held camcorder. Using only that videotape, is it possible to create a three-dimensional model of the room as well as determine the camera trajectory?

The solution to this problem, often called the structure-from-motion problem, has eluded vision researchers for years. We have developed a new method, called Factorization, which can give a robust solution to this problem. The method is based on the theorem that the geometrical constraints due to incidence relations among projection rays can be expressed as the degeneracy of a matrix that gathers all the image measurements. The theorem results in an algorithm that factorizes the measurement matrix into two matrices that represent shape and motion, respectively, based on the robust singular value decomposition (SVD) technique.

Personnel [Past Members]

	Name	Title	Email Address
	Takeo Kanade	U.A. and Helen Whitaker University Prof., RI/CS	tk@cs.cmu.edu

Recent publications [View all 15 publications]

• Gauge Fixing for Accurate 3D Estimation
  D.D. Morris, K. Kanatani, and T. Kanade
  Computer Vision and Pattern Recognition, December, 2001. [Abstract]
  Download: pdf [284 KB], ps.gz [1438 KB] copyrighted

• Gauges and gauge transformations for uncertainty description of geometric structure with indeterminacy
  K. Kanatani and D.D. Morris
  IEEE Transactions on Information Theory, Vol. 47, No. 5, July, 2001, pp. 2017--2028. [Abstract]
  Download: pdf [476 KB], ps.gz [1076 KB] copyrighted

• Gauge Freedoms and Uncertainty Modeling for 3D Computer Vision
  D.D. Morris
  doctoral dissertation, tech. report CMU-RI-TR-01-15, Robotics Institute, Carnegie Mellon University, March, 2001. [Abstract]
  Download: pdf [12780 KB] copyrighted

• 3D Model Accuracy and Gauge Fixing
  D.D. Morris, K. Kanatani, and T. Kanade
  tech. report CMU-RI-TR-00-32, Robotics Institute, Carnegie Mellon University, December, 2000. [Abstract]
  Download: pdf [337 KB], ps.gz [1772 KB] copyrighted

• Creating 3D Models with Uncalibrated Cameras
  M. Han and T. Kanade
  proceeding of IEEE Computer Society Workshop on the Application of Computer Vision (WACV2000), December, 2000. [Abstract]
  Download: pdf [294 KB] copyrighted

• Image-Consistent Surface Triangulation
  D.D. Morris and T. Kanade
  Computer Vision and Pattern Recognition (CVPR 2000), IEEE Computer Society, Vol. 1, June, 2000, pp. 332-338. [Abstract]
  Download: pdf [1180 KB], ps.gz [4793 KB] copyrighted

• Reconstruction of a Scene with Multiple Linearly Moving Objects
  M. Han and T. Kanade
  Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR '00), June, 2000. [Abstract]
  Download: pdf [262 KB], ps.gz [1173 KB] copyrighted

• Gauges and gauge transformations in 3-D reconstruction from a sequence of images
  K. Kanatani and D.D. Morris
  Proceedings of Fourth Asian Conference on Computer Vision (ACCV 2000), 2000, pp. 1046-1051. [Abstract]
  Download: pdf [240 KB], ps.gz [100 KB] copyrighted

• Uncertainty Modeling for Optimal Structure from Motion
  D.D. Morris, K. Kanatani, and T. Kanade
  Vision Algorithms Theory and Practice, Springer LNCS, September, 1999. [Abstract]
  Download: pdf [366 KB], ps.gz [661 KB] copyrighted

• Interactive Construction of 3D Models from Panoramic Mosaics
  H. Shum, M. Han, and R. Szeliski
  Proc. of CVPR'98, June, 1998. [Abstract]
  Download: pdf [289 KB], ps.gz [2681 KB] copyrighted