In this paper, we propose a novel framework to jointly recover the illumination environment and an estimate of the cast shadows in a scene from a single image, given coarse 3D geo...
In this paper, we present a novel localized Markov random field (MRF) method based on superpixels for region segmentation. Early vision problems could be formulated as pixel label...
Markov Random Fields (MRFs) are proposed as viable stochastic models for the spatial distribution of gray level intensities for images of human faces. These models are trained usi...
Consider the problem of joint parameter estimation and prediction in a Markov random field: i.e., the model parameters are estimated on the basis of an initial set of data, and th...
In real world, a scene is composed by many characteristics. Intrinsic images represent these characteristics by two components, reflectance (the albedo of each point) and shading (...