Fronts propagating with curvature-dependent speed: algorithms based on Hamilton-Jacobi formulations
Journal of Computational Physics
CVGIP: Image Understanding
IEEE Transactions on Pattern Analysis and Machine Intelligence
Numerical methods for shape-from-shading: A new survey with benchmarks
Computer Vision and Image Understanding
Hi-index | 0.00 |
This paper extends the MRF formulation approach [6] for solving the shape from shading problem. Our method uses simulated annealing Gibbs sampling to minimize a lattice energy for an MRF formulation which characterizes the Shape from Shading (SFS) problem under Lambertian reflectance conditions. We explore a spatial multi-resolution approach which refines the digital elevation map of the surface in a coarse to fine manner. This framework allows us to deal with images as projections rather than affine mappings, and allows us to incorporate a local occlusion constraint to cater for more general scene/lighting configurations.