Share on Facebook Tweet on Twitter Share on LinkedIn Share by email
Efficient Object Category Recognition using Classemes

Lorenzo Torresani, Martin Szummer, and Andrew Fitzgibbon

Abstract

We introduce a new descriptor for images which allows the construction of efficient and compact classifiers with good accuracy on object category recognition. The descriptor is the output of a large number of weakly trained object category classifiers on the image. The trained categories are selected from an ontology of visual concepts, but the intention is not to encode an explicit decomposition of the scene. Rather, we accept that existing object category classifiers often encode not the category per se but ancillary image characteristics; and that these ancillary characteristics can combine to represent visual classes unrelated to the constituent categories' semantic meanings.

The advantage of this descriptor is that it allows object-category queries to be made against image databases using efficient classifiers (efficient at test time) such as linear support vector machines, and allows these queries to be for novel categories. Even when the representation is reduced to 200 bytes per image, classification accuracy on object category recognition is comparable with the state of the art (36% versus 42%), but at orders of magnitude lower computational cost.

Details

Publication typeInproceedings
Published inEuropean Conference on Computer Vision (ECCV)
Pages776–789
SeriesLNCS 6311
PublisherSpringer Verlag
> Publications > Efficient Object Category Recognition using Classemes