Projective Reconstruction with Missing Data Combining 2D Reprojection Error and Subspace Method

Factorization algorithms widely used in projective reconstruction handle all of images uniformly without preferential treatment for any image. However, they require that all the points are visible in all views. Due to reasons, such as occlusion, the demand can not be satisfied. For the factorization method to be better applicable, we propose a new factorization-based algorithm combining 2D reprojection error and subspace method for projective reconstruction with missing data, which estimates projective shape, projection matrices, projective depths and missing data iteratively. Estimation problems of projective shape and projection matrices are formulated in terms of the minimization of 2D reprojection error. The subspace method is used to estimate projective depths. Experimental results using both synthetic data and real images are provided to illustrate the performance of the proposed method.