A Bayesian Hierarchical Model for Learning Natural Scene Categories
2005Vol. 2, pp. 524–531
Citations Over TimeTop 1% of 2005 papers
Abstract
We propose a novel approach to learn and recognize natural scene categories. Unlike previous work, it does not require experts to annotate the training set. We represent the image of a scene by a collection of local regions, denoted as codewords obtained by unsupervised learning. Each region is represented as part of a "theme". In previous work, such themes were learnt from hand-annotations of experts, while our method learns the theme distributions as well as the codewords distribution over the themes without supervision. We report satisfactory categorization performances on a large set of 13 categories of complex scenes.
Related Papers
- → Visual Search and the Collapse of Categorization.(2005)35 cited
- → The neurobiology of categorization(2010)4 cited
- → Two categorization patterns in idiom semantics(2016)1 cited
- On the Reasons for Cognitive Differences During Categorization(2009)
- → Is one object enough? Diagnosticity of single objects for fast scene categorization(2022)