Graph fusion using global descriptors for image retrieval. Moraga, C. R., Mardones, T., & Allende, H. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), volume 9423, pages 290-297, 2015. doi abstract bibtex © Springer International Publishing Switzerland 2015.This paper addresses the problem of content-based image retrieval in a large-scale setting. Recently several graph-based image retrieval systems to fuse different representations have been proposed with excellent results, however most of them use at least one representation based on local descriptors that does not scale very well with the number the images, hurting time and memory requirements as the database grows. This motivated us to investigate the possibility to retain the performance of local descriptor methods while using only global descriptions of the image. Thus, we propose a graph-based query fusion approach -where we combine several representations based on aggregating local descriptors such as Fisher Vectors- using distance and neighborhood information to evaluate the individual importance of each element in every query. Performance is analyzed in different time and memory constrained scenarios. Experiments are performed on 3 public datasets: the UKBench, Holidays and MIRFLICKR-1M, obtaining state of the art performance.
@inproceedings{10.1007/978-3-319-25751-8_35,
abstract = "© Springer International Publishing Switzerland 2015.This paper addresses the problem of content-based image retrieval in a large-scale setting. Recently several graph-based image retrieval systems to fuse different representations have been proposed with excellent results, however most of them use at least one representation based on local descriptors that does not scale very well with the number the images, hurting time and memory requirements as the database grows. This motivated us to investigate the possibility to retain the performance of local descriptor methods while using only global descriptions of the image. Thus, we propose a graph-based query fusion approach -where we combine several representations based on aggregating local descriptors such as Fisher Vectors- using distance and neighborhood information to evaluate the individual importance of each element in every query. Performance is analyzed in different time and memory constrained scenarios. Experiments are performed on 3 public datasets: the UKBench, Holidays and MIRFLICKR-1M, obtaining state of the art performance.",
year = "2015",
title = "Graph fusion using global descriptors for image retrieval",
volume = "9423",
keywords = "Fisher vector , Global descriptors , Graph fusion , Large scale image retrieval",
pages = "290-297",
doi = "10.1007/978-3-319-25751-8\_35",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
author = "Moraga, Claudio R. and Mardones, Tomás and Allende, Héctor"
}
Downloads: 0
{"_id":"tiMGKfGkL9PBNqNu7","bibbaseid":"moraga-mardones-allende-graphfusionusingglobaldescriptorsforimageretrieval-2015","downloads":0,"creationDate":"2017-04-03T15:16:25.288Z","title":"Graph fusion using global descriptors for image retrieval","author_short":["Moraga, C. R.","Mardones, T.","Allende, H."],"year":2015,"bibtype":"inproceedings","biburl":"https://1fichier.com/?yjj84y68k0","bibdata":{"bibtype":"inproceedings","type":"inproceedings","abstract":"© Springer International Publishing Switzerland 2015.This paper addresses the problem of content-based image retrieval in a large-scale setting. Recently several graph-based image retrieval systems to fuse different representations have been proposed with excellent results, however most of them use at least one representation based on local descriptors that does not scale very well with the number the images, hurting time and memory requirements as the database grows. This motivated us to investigate the possibility to retain the performance of local descriptor methods while using only global descriptions of the image. Thus, we propose a graph-based query fusion approach -where we combine several representations based on aggregating local descriptors such as Fisher Vectors- using distance and neighborhood information to evaluate the individual importance of each element in every query. Performance is analyzed in different time and memory constrained scenarios. Experiments are performed on 3 public datasets: the UKBench, Holidays and MIRFLICKR-1M, obtaining state of the art performance.","year":"2015","title":"Graph fusion using global descriptors for image retrieval","volume":"9423","keywords":"Fisher vector , Global descriptors , Graph fusion , Large scale image retrieval","pages":"290-297","doi":"10.1007/978-3-319-25751-8_35","booktitle":"Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)","author":[{"propositions":[],"lastnames":["Moraga"],"firstnames":["Claudio","R."],"suffixes":[]},{"propositions":[],"lastnames":["Mardones"],"firstnames":["Tomás"],"suffixes":[]},{"propositions":[],"lastnames":["Allende"],"firstnames":["Héctor"],"suffixes":[]}],"bibtex":"@inproceedings{10.1007/978-3-319-25751-8_35,\n abstract = \"© Springer International Publishing Switzerland 2015.This paper addresses the problem of content-based image retrieval in a large-scale setting. Recently several graph-based image retrieval systems to fuse different representations have been proposed with excellent results, however most of them use at least one representation based on local descriptors that does not scale very well with the number the images, hurting time and memory requirements as the database grows. This motivated us to investigate the possibility to retain the performance of local descriptor methods while using only global descriptions of the image. Thus, we propose a graph-based query fusion approach -where we combine several representations based on aggregating local descriptors such as Fisher Vectors- using distance and neighborhood information to evaluate the individual importance of each element in every query. Performance is analyzed in different time and memory constrained scenarios. Experiments are performed on 3 public datasets: the UKBench, Holidays and MIRFLICKR-1M, obtaining state of the art performance.\",\n year = \"2015\",\n title = \"Graph fusion using global descriptors for image retrieval\",\n volume = \"9423\",\n keywords = \"Fisher vector , Global descriptors , Graph fusion , Large scale image retrieval\",\n pages = \"290-297\",\n doi = \"10.1007/978-3-319-25751-8\\_35\",\n booktitle = \"Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)\",\n author = \"Moraga, Claudio R. and Mardones, Tomás and Allende, Héctor\"\n}\n\n","author_short":["Moraga, C. R.","Mardones, T.","Allende, H."],"key":"10.1007/978-3-319-25751-8_35","id":"10.1007/978-3-319-25751-8_35","bibbaseid":"moraga-mardones-allende-graphfusionusingglobaldescriptorsforimageretrieval-2015","role":"author","urls":{},"keyword":["Fisher vector","Global descriptors","Graph fusion","Large scale image retrieval"],"downloads":0},"search_terms":["graph","fusion","using","global","descriptors","image","retrieval","moraga","mardones","allende"],"keywords":["fisher vector","global descriptors","graph fusion","large scale image retrieval"],"authorIDs":[],"dataSources":["mqyNxb9twi6rYAXFv"]}