Statistical Learning of Visual Feature Hierarchies

We propose an unsupervised, probabilistic method for learning visual feature hierarchies. Starting from local, low-level features computed at interest point locations, the method combines these primitives into high-level abstractions. Our appearance-based learning method uses local statistical analysis between features and Expectation- Maximization (EM) to identify and code spatial correlations. Spatial correlation is asserted when two features tend to occur at the same relative position of each other. This learning scheme results in a graphical model that allows a probabilistic representation of a flexible visual feature hierarchy. For feature detection, evidence is propagated using Nonparametric Belief Propagation (NBP), a recent generalization of particle filtering. In experiments, the proposed approach demonstrates efficient learning and robust detection of object models in the presence of clutter and occlusion and under view point changes.