Stereo registration using kernel density correlation

A common approach to solving the stereo registration problem is to model the disparity function as a discrete-valued Markov Random Field. The key problems with this approach are its combinatorial computational complexity, and the discretization of the obtained disparity estimates. In this paper, we propose a framework that addresses the requirements of a robust continuous domain formulation for stereo registration. The proposed formulation is based on a new measure, derived from the correlation of empirical probability density distributions estimated using kernel estimators. We term this the kernel density correlation (KDC) measure. The proposed framework takes the form of an energy minimization formulation which is efficiently solved using the technique of variational optimization. We prove the convergence properties of the resultant iterative algorithm, and compare the performance of the proposed formulation to that of a state-of-the-art stereo registration approach.