Reverse-rendering method for digital modeling

Patent 6990230 Issued on January 24, 2006. Estimated Expiration Date:

November 17, 2023. Estimated Expiration Date is calculated based on simple USPTO term provisions. It does not account for terminal disclaimers, term adjustments, failure to pay maintenance fees, or other factors which might affect the term of a patent.

Abstract Claims Description Full Text

Patent References

Position-and-attitude recognition method and apparatus by use of image pickup means
Patent #: 5499306
Issued on: 03/12/1996
Inventor: Sasaki, et al.

Error-bounded antialiased rendering of complex scenes
Patent #: 5600763
Issued on: 02/04/1997
Inventor: Greene, et al.

Apparatus and method for recreating and manipulating a 3D object based on a 2D projection thereof
Patent #: 5821943
Issued on: 10/13/1998
Inventor: Shashua

System and method for directly estimating three-dimensional structure of objects in a scene and camera motion from three two-dimensional views of the scene
Patent #: 6052124
Issued on: 04/18/2000
Inventor: Stein, et al.

Method and apparatus for reconstructing geometry using geometrically constrained structure from motion with points on planes Patent #: 6137491
Issued on: 10/24/2000
Inventor: Szeliski

Inventor

Piponi, Daniele Paolo David

Assignee

Warner Bros. Entertainment Inc.

Application

No. 10715869 filed on 11/17/2003

US Classes:

382/154, 3-D or stereo imaging analysis345/419, Three-dimension382/285Mapping 2-D image onto a 3-D surface

Examiners

Primary: Couso, Jose L.
Assistant: Lu, Tom Y.

Attorney, Agent or Firm

O'Melveny & Myers LLP

International Class

G06K 9/00

Abstract

A method for automatically or semi-automatically constructing a digital 3D model of a scene from photographic data and photogrammetry data includes defining an initial rough model as a solution estimate. A reverse rendering step includes a second-order solution method that employs automatic differentiation techniques to accurately compute derivatives of an error function. In an embodiment of the method, at least one camera is placed within the scene being constructed, and photographic data from this camera is used in the solution process.

Other References

Frank Dellaert, “Monte Carlo EM for Data-Association and its Applications in Computer Vision”, Sep. 21, 2001, CMU-CS-01-153, School of Computer Science, Carnegie Mellon University.
“Monte Carlo EM for Data-Association and its Applications in Computer Vision” by Frank Dellaert, School of Computer Science; Carnegie Mellon University, Pittsburg, PA 15213; CMU-CS-01-153; Sep. 21, 2001.
“Bundle Adjustment—A Modern Synthesis” by Bill Trigg et al., INRIA Rhine-Alpes, 655 avenue de L'Europe, 38330 Montbonnot, France; (pp. 1-71).
“Automatic Differentiation: Obtaining Fast and Reliable Derivatives—Fast” by Christian H. Bischof, Alan Carle, Peyvand M. Khademi and Gordon Pusch, Argonne Preprint MCS-P484-1194; Also to appear in Proc. of the SIAM Symposium on Control Problems in Industry San Diego, Jul. 1994; (pp. 1-17).
“A Collection of Automatic Differentiation Tools”, http://www-unix.mcs.anl.gov/autodiff/AD_—Tools.
“On Automatic Differentiation” by Andreas Griewank, Argonne National Laboratory; Mathematics and Computer Science Division, Preprint ANL/MCS-P-10-1088; Nov. 1988.