Semantic Kernel Forests from Multiple Taxonomies
Sung Ju Hwang, Kristen Grauman, Fei Sha · 2012
We propose a discriminative feature learning approach that leverages multiple hi-erarchical taxonomies representing different semantic views. For each taxonomy, we first learn a tree of semantic kernels, where each node has a Mahalanobis ker-nel optimized to distinguish between the classes in its children nodes. Then, using the resulting semantic kernel forest, we learn class-specific kernel combinations to select only those kernels relevant for category recognition, with a novel hier-archical regularizer that exploits the taxonomies ’ structure. We demonstrate our method on challenging object recognition datasets. 1