The Community for Technology Leaders
2013 IEEE Conference on Computer Vision and Pattern Recognition (2009)
Miami, FL, USA
June 20, 2009 to June 25, 2009
ISBN: 978-1-4244-3992-8
pp: 1169-1176
H. Jegou , LJK, INRIA Grenoble, Grenoble, France
C. Schmid , LJK, INRIA Grenoble, Grenoble, France
M. Douze , LJK, INRIA Grenoble, Grenoble, France
ABSTRACT
Burstiness, a phenomenon initially observed in text retrieval, is the property that a given visual element appears more times in an image than a statistically independent model would predict. In the context of image search, burstiness corrupts the visual similarity measure, i.e., the scores used to rank the images. In this paper, we propose a strategy to handle visual bursts for bag-of-features based image search systems. Experimental results on three reference datasets show that our method significantly and consistently outperforms the state of the art.
INDEX TERMS
reference datasets, visual elements, burstiness, text retrieval, visual bursts, image search systems
CITATION
H. Jegou, C. Schmid, M. Douze, "On the burstiness of visual elements", 2013 IEEE Conference on Computer Vision and Pattern Recognition, vol. 00, no. , pp. 1169-1176, 2009, doi:10.1109/CVPRW.2009.5206609
515 ms
(Ver 3.3 (11022016))