Noise in bilinear problems

Despite the wide application of bilinear problems to problems both in computer vision and in other fields, their behaviour under the effects of noise is still poorly understood. In this paper, we show analytically that marginal distributions on the solution components of a bilinear problem can be bimodal, even with Gaussian measurement error. We demonstrate and compare three different methods of estimating the covariance of a solution. We show that the Hessian at the mode substantially underestimates covariance. Many problems in computer vision can be posed as bilinear problems: i.e. one must find a solution to a set of equations of the form.

[1]  Gordon M. Crippen,et al.  Distance Geometry and Molecular Conformation , 1988 .

[2]  Takeo Kanade,et al.  A unified factorization algorithm for points, line segments and planes with uncertainty models , 1998, Sixth International Conference on Computer Vision (IEEE Cat. No.98CH36271).

[3]  David A. Forsyth,et al.  Bayesian structure from motion , 1999, Proceedings of the Seventh IEEE International Conference on Computer Vision.

[4]  Tim B. Swartz,et al.  Approximating Integrals Via Monte Carlo and Deterministic Methods , 2000 .

[5]  Philip F. McLauchlan,et al.  Gauge Independence in Optimization Algorithms for 3D Vision , 1999, Workshop on Vision Algorithms.