Supervised Learning of Semantic Classes for Image Annotation and Retrieval

TLDR

The paper proposes a probabilistic formulation for semantic image annotation and retrieval. The method treats annotation and retrieval as classification, representing images as bags of localized features, estimating mixture densities per image, pooling these into class densities via a hierarchical EM algorithm justified by multiple‑instance learning. Experiments show the supervised formulation achieves lower error, higher accuracy than prior methods, is computationally efficient, and remains robust to parameter choices.

Abstract

A probabilistic formulation for semantic image annotation and retrieval is proposed. Annotation and retrieval are posed as classification problems where each class is defined as the group of database images labeled with a common semantic label. It is shown that, by establishing this one-to-one correspondence between semantic labels and semantic classes, a minimum probability of error annotation and retrieval are feasible with algorithms that are 1) conceptually simple, 2) computationally efficient, and 3) do not require prior semantic segmentation of training images. In particular, images are represented as bags of localized feature vectors, a mixture density estimated for each image, and the mixtures associated with all images annotated with a common semantic label pooled into a density estimate for the corresponding semantic class. This pooling is justified by a multiple instance learning argument and performed efficiently with a hierarchical extension of expectation-maximization. The benefits of the supervised formulation over the more complex, and currently popular, joint modeling of semantic label and visual feature distributions are illustrated through theoretical arguments and extensive experiments. The supervised formulation is shown to achieve higher accuracy than various previously published methods at a fraction of their computational cost. Finally, the proposed method is shown to be fairly robust to parameter tuning.