Beyond bags of features: Spatial pyramid matching for recognizing natural scene categories

DMCA

Beyond bags of features: Spatial pyramid matching for recognizing natural scene categories

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Cordelia Schmid

Venue:	In CVPR
Citations:	1858 - 47 self

BibTeX

@INPROCEEDINGS{Schmid_beyondbags,
    author = {Cordelia Schmid},
    title = {Beyond bags of features: Spatial pyramid matching for recognizing natural scene categories},
    booktitle = {In CVPR},
    year = {}
}

OpenURL

Abstract

This paper presents a method for recognizing scene categories based on approximate global geometric correspondence. This technique works by partitioning the image into increasingly fine sub-regions and computing histograms of local features found inside each sub-region. The resulting “spatial pyramid ” is a simple and computationally efficient extension of an orderless bag-of-features image representation, and it shows significantly improved performance on challenging scene categorization tasks. Specifically, our proposed method exceeds the state of the art on the Caltech-101 database and achieves high accuracy on a large database of fifteen natural scene categories. The spatial pyramid framework also offers insights into the success of several recently proposed image descriptions, including Torralba’s “gist ” and Lowe’s SIFT descriptors. 1.

Keyphrases

spatial pyramid beyond bag natural scene category achieves high accuracy sift descriptor caltech-101 database large database torralba gist scene category spatial pyramid framework efficient extension orderless bag-of-features image representation local feature image description fine sub-regions approximate global geometric correspondence fifteen natural scene category scene categorization task

DMCA