Joint Inverted Indexing. Xia, Y., He, K., Wen, F., & Sun, J. In 2013 IEEE International Conference on Computer Vision, pages 3416–3423, December, 2013. ISSN: 2380-7504
doi  abstract   bibtex   
Inverted indexing is a popular non-exhaustive solution to large scale search. An inverted file is built by a quantizer such as k-means or a tree structure. It has been found that multiple inverted files, obtained by multiple independent random quantizers, are able to achieve practically good recall and speed. Instead of computing the multiple quantizers independently, we present a method that creates them jointly. Our method jointly optimizes all code words in all quantizers. Then it assigns these code words to the quantizers. In experiments this method shows significant improvement over various existing methods that use multiple independent quantizers. On the one-billion set of SIFT vectors, our method is faster and more accurate than a recent state-of-the-art inverted indexing method.
@inproceedings{xia_joint_2013,
	title = {Joint {Inverted} {Indexing}},
	doi = {10.1109/ICCV.2013.424},
	abstract = {Inverted indexing is a popular non-exhaustive solution to large scale search. An inverted file is built by a quantizer such as k-means or a tree structure. It has been found that multiple inverted files, obtained by multiple independent random quantizers, are able to achieve practically good recall and speed. Instead of computing the multiple quantizers independently, we present a method that creates them jointly. Our method jointly optimizes all code words in all quantizers. Then it assigns these code words to the quantizers. In experiments this method shows significant improvement over various existing methods that use multiple independent quantizers. On the one-billion set of SIFT vectors, our method is faster and more accurate than a recent state-of-the-art inverted indexing method.},
	language = {en},
	booktitle = {2013 {IEEE} {International} {Conference} on {Computer} {Vision}},
	author = {Xia, Yan and He, Kaiming and Wen, Fang and Sun, Jian},
	month = dec,
	year = {2013},
	note = {ISSN: 2380-7504},
	keywords = {\#Analysis, \#ICCV{\textgreater}13, \#Joint, \#Machine Learning, \#Vision, /unread, Indexing, Joints, Lattices, Optimization, Quantization (signal), Search engines, Vectors},
	pages = {3416--3423},
}

Downloads: 0