Unified Computation of Strict Maximum Likelihood for Geometric Fitting
Access this Article
Search this Article
A new numerical scheme is presented for computing strict maximum likelihood (ML) of geometricfitting problems having an implicit constraint. Our approach is orthogonal projection of observationsonto a parameterized surface defined by the constraint. Assuming a linearly separable nonlinear constraint, we show that a theoretically global solution can be obtained by iterative Sampson error minimization. Our approach is illustrated by ellipse fitting and fundamental matrix computation. Our method also encompasses optimal correction, computing, e.g., perpendiculars to an ellipse and triangulating stereo images. A detailed discussion is given to technical and practical issues about our approach.
- Memoirs of the Faculty of Engineering, Okayama University
Memoirs of the Faculty of Engineering, Okayama University (44), 13-23, 2010-01
Faculty of Engineering, Okayama University