Abstract

We seek to discover the object categories depicted in a set of unlabelled images. We achieve this using a model developed in the statistical text literature: probabilistic Latent Semantic Analysis (pLSA). In text analysis this is used to discover topics in a corpus using the bag-of-words document representation. Here we treat object categories as topics, so that an image containing instances of several categories is modeled as a mixture of topics. The model is applied to images by using a visual analogue of a word, formed by vector quantizing SIFT-like region descriptors. The topic discovery approach successfully translates to the visual domain: for a small set of objects, we show that both the object categories and their approximate spatial layout are found without supervision. Performance of this unsupervised method is compared to the supervised approach of Fergus et al. [8] on a set of unseen images containing only one object per image. We also extend the bag-of-words vocabulary to include ‘doublets’ which encode spatially local co-occurring regions. It is demonstrated that this extended vocabulary gives a cleaner image segmentation. Finally, the classification and segmentation methods are applied to a set of images containing multiple objects per image. These results demonstrate that we can successfully build object class models from an unsupervised analysis of images.

Keywords

Artificial intelligenceComputer scienceProbabilistic latent semantic analysisPattern recognition (psychology)VocabularyObject (grammar)Set (abstract data type)Bag-of-words model in computer visionSegmentationRepresentation (politics)Bag-of-words modelImage segmentationClass (philosophy)Image (mathematics)Scale-invariant feature transformProbabilistic logicWord (group theory)Visual WordNatural language processingImage retrievalMathematics

Affiliated Institutions

Related Publications

Publication Info

Year
2005
Type
article
Pages
370-377 Vol. 1
Citations
980
Access
Closed

External Links

Social Impact

Social media, news, blog, policy document mentions

Citation Metrics

980
OpenAlex

Cite This

Josef Šivic, Bryan Russell, Alexei A. Efros et al. (2005). Discovering objects and their location in images. , 370-377 Vol. 1. https://doi.org/10.1109/iccv.2005.77

Identifiers

DOI
10.1109/iccv.2005.77