Conditional random field with the multi-granular contextual information for pixel labeling

To make full use of the contextual information object recognition and scene understanding, a multi-granular context conditional random field (MGCCRF) model is presented to combine context information in a variety of scales. It is efficiently implemented through extending the pairwise clique to the multi-granular context windows. In the fine-granular context window, the label consistency of similar features can be obtained with the probability of the label transferring between two adjacent pixels. At the same time, the spatial relationships among different classes in the coarse-granular context window are explicated in details. To train the MGCCRF model, a piecewise training method with the bound optimization algorithm is designed to improve the performance. Experiments on two real-world image databases show that compared with other methods, the modified conditional random field model is more competitive and effective in terms of the quantitative and qualitative labeling performance.