Aggregating Local Image Descriptors into Compact Codes

Aggregating Local Image Descriptors into Compact Codes. Jegou, H., Perronnin, F., Douze, M., Sanchez, J., Perez, P., & Schmid, C. IEEE Transactions on Pattern Analysis and Machine Intelligence, 34(9):1704–1716, September, 2012.

Paper doi abstract bibtex

This paper addresses the problem of large-scale image search. Three constraints have to be taken into account: search accuracy, efﬁciency, and memory usage. We ﬁrst present and evaluate different ways of aggregating local image descriptors into a vector and show that the Fisher kernel achieves better performance than the reference bag-of-visual words approach for any given vector dimension. We then jointly optimize dimensionality reduction and indexing in order to obtain a precise vector comparison as well as a compact representation. The evaluation shows that the image representation can be reduced to a few dozen bytes while preserving high accuracy. Searching a 100 million image dataset takes about 250 ms on one processor core.

@article{jegou_aggregating_2012,
	title = {Aggregating {Local} {Image} {Descriptors} into {Compact} {Codes}},
	volume = {34},
	issn = {0162-8828, 2160-9292},
	url = {http://ieeexplore.ieee.org/document/6104058/},
	doi = {10.1109/TPAMI.2011.235},
	abstract = {This paper addresses the problem of large-scale image search. Three constraints have to be taken into account: search accuracy, efﬁciency, and memory usage. We ﬁrst present and evaluate different ways of aggregating local image descriptors into a vector and show that the Fisher kernel achieves better performance than the reference bag-of-visual words approach for any given vector dimension. We then jointly optimize dimensionality reduction and indexing in order to obtain a precise vector comparison as well as a compact representation. The evaluation shows that the image representation can be reduced to a few dozen bytes while preserving high accuracy. Searching a 100 million image dataset takes about 250 ms on one processor core.},
	language = {en},
	number = {9},
	urldate = {2022-03-02},
	journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
	author = {Jegou, H. and Perronnin, F. and Douze, M. and Sanchez, J. and Perez, P. and Schmid, C.},
	month = sep,
	year = {2012},
	pages = {1704--1716},
}

Downloads: 0

{"_id":"gFfdaqMBhZwRWt9ni","bibbaseid":"jegou-perronnin-douze-sanchez-perez-schmid-aggregatinglocalimagedescriptorsintocompactcodes-2012","author_short":["Jegou, H.","Perronnin, F.","Douze, M.","Sanchez, J.","Perez, P.","Schmid, C."],"bibdata":{"bibtype":"article","type":"article","title":"Aggregating Local Image Descriptors into Compact Codes","volume":"34","issn":"0162-8828, 2160-9292","url":"http://ieeexplore.ieee.org/document/6104058/","doi":"10.1109/TPAMI.2011.235","abstract":"This paper addresses the problem of large-scale image search. Three constraints have to be taken into account: search accuracy, efﬁciency, and memory usage. We ﬁrst present and evaluate different ways of aggregating local image descriptors into a vector and show that the Fisher kernel achieves better performance than the reference bag-of-visual words approach for any given vector dimension. We then jointly optimize dimensionality reduction and indexing in order to obtain a precise vector comparison as well as a compact representation. The evaluation shows that the image representation can be reduced to a few dozen bytes while preserving high accuracy. Searching a 100 million image dataset takes about 250 ms on one processor core.","language":"en","number":"9","urldate":"2022-03-02","journal":"IEEE Transactions on Pattern Analysis and Machine Intelligence","author":[{"propositions":[],"lastnames":["Jegou"],"firstnames":["H."],"suffixes":[]},{"propositions":[],"lastnames":["Perronnin"],"firstnames":["F."],"suffixes":[]},{"propositions":[],"lastnames":["Douze"],"firstnames":["M."],"suffixes":[]},{"propositions":[],"lastnames":["Sanchez"],"firstnames":["J."],"suffixes":[]},{"propositions":[],"lastnames":["Perez"],"firstnames":["P."],"suffixes":[]},{"propositions":[],"lastnames":["Schmid"],"firstnames":["C."],"suffixes":[]}],"month":"September","year":"2012","pages":"1704–1716","bibtex":"@article{jegou_aggregating_2012,\n\ttitle = {Aggregating {Local} {Image} {Descriptors} into {Compact} {Codes}},\n\tvolume = {34},\n\tissn = {0162-8828, 2160-9292},\n\turl = {http://ieeexplore.ieee.org/document/6104058/},\n\tdoi = {10.1109/TPAMI.2011.235},\n\tabstract = {This paper addresses the problem of large-scale image search. Three constraints have to be taken into account: search accuracy, efﬁciency, and memory usage. We ﬁrst present and evaluate different ways of aggregating local image descriptors into a vector and show that the Fisher kernel achieves better performance than the reference bag-of-visual words approach for any given vector dimension. We then jointly optimize dimensionality reduction and indexing in order to obtain a precise vector comparison as well as a compact representation. The evaluation shows that the image representation can be reduced to a few dozen bytes while preserving high accuracy. Searching a 100 million image dataset takes about 250 ms on one processor core.},\n\tlanguage = {en},\n\tnumber = {9},\n\turldate = {2022-03-02},\n\tjournal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},\n\tauthor = {Jegou, H. and Perronnin, F. and Douze, M. and Sanchez, J. and Perez, P. and Schmid, C.},\n\tmonth = sep,\n\tyear = {2012},\n\tpages = {1704--1716},\n}\n\n","author_short":["Jegou, H.","Perronnin, F.","Douze, M.","Sanchez, J.","Perez, P.","Schmid, C."],"key":"jegou_aggregating_2012","id":"jegou_aggregating_2012","bibbaseid":"jegou-perronnin-douze-sanchez-perez-schmid-aggregatinglocalimagedescriptorsintocompactcodes-2012","role":"author","urls":{"Paper":"http://ieeexplore.ieee.org/document/6104058/"},"metadata":{"authorlinks":{}},"html":""},"bibtype":"article","biburl":"https://bibbase.org/zotero/mxmplx","dataSources":["aXmRAq63YsH7a3ufx"],"keywords":[],"search_terms":["aggregating","local","image","descriptors","compact","codes","jegou","perronnin","douze","sanchez","perez","schmid"],"title":"Aggregating Local Image Descriptors into Compact Codes","year":2012}