Salient Object Detection: A Benchmark

Salient Object Detection: A Benchmark. Borji, A., Sihite, D. N., & Itti, L. In Proc. European Conference on Computer Vision (ECCV), Florence, Italy (LNCS 7573), pages 414-429, Oct, 2012.
abstract bibtex

Several salient object detection approaches have been published which have been assessed using different evaluation scores and datasets resulting in discrepancy in model comparison. This calls for a methodological framework to compare existing models and evaluate their pros and cons. We analyze benchmark datasets and scoring techniques and, for the first time, provide a quantitative comparison of 35 state-of-the-art saliency detection models. We find that some models perform consistently better than the others. Saliency models that intend to predict eye fixations perform lower on segmentation datasets compared to salient object detection algorithms. Further, we propose combined models which show that integration of the few best models outperforms all models over other datasets. By analyzing the consistency among the best models and among humans for each scene, we identify the scenes where models or humans fail to detect the most salient object. We highlight the current issues and propose future research directions.

@inproceedings{ Borji_etal12eccv,
  author = {A. Borji and D. N. Sihite and L. Itti},
  title = {Salient Object Detection: A Benchmark},
  booktitle = {Proc. European Conference on Computer Vision (ECCV), Florence, Italy (LNCS 7573)},
  abstract = {Several salient object detection approaches have been
                  published which have been assessed using different
                  evaluation scores and datasets resulting in
                  discrepancy in model comparison. This calls for a
                  methodological framework to compare existing models
                  and evaluate their pros and cons. We analyze
                  benchmark datasets and scoring techniques and, for
                  the first time, provide a quantitative comparison of
                  35 state-of-the-art saliency detection models. We
                  find that some models perform consistently better
                  than the others. Saliency models that intend to
                  predict eye fixations perform lower on segmentation
                  datasets compared to salient object detection
                  algorithms. Further, we propose combined models
                  which show that integration of the few best models
                  outperforms all models over other datasets. By
                  analyzing the consistency among the best models and
                  among humans for each scene, we identify the scenes
                  where models or humans fail to detect the most
                  salient object. We highlight the current issues and
                  propose future research directions.},
  month = {Oct},
  year = {2012},
  pages = {414-429},
  review = {full/conf},
  type = {bu;td;mod;cv},
  if = {2012 acceptance rate: 25.0%},
  file = {http://ilab.usc.edu/publications/doc/Borji_etal12eccv.pdf}
}

Downloads: 0

{"_id":{"_str":"5298a19f9eb585cc260007c0"},"__v":0,"authorIDs":[],"author_short":["Borji, A.","Sihite, D.<nbsp>N.","Itti, L."],"bibbaseid":"borji-sihite-itti-salientobjectdetectionabenchmark-2012","bibdata":{"html":"<div class=\"bibbase_paper\"> \n\n\n<span class=\"bibbase_paper_titleauthoryear\">\n\t<span class=\"bibbase_paper_title\"><a name=\"Borji_etal12eccv\"> </a>Salient Object Detection: A Benchmark.</span>\n\t<span class=\"bibbase_paper_author\">\nBorji, A.; Sihite, D. N.; and Itti, L.</span>\n\t\n</span>\n\n\n\nIn\n<i>Proc. European Conference on Computer Vision (ECCV), Florence, Italy (LNCS 7573)</i>, page 414-429, Oct 2012.\n\n\n\n\n\n<br class=\"bibbase_paper_content\"/>\n\n<span class=\"bibbase_paper_content\">\n \n \n \n <a href=\"javascript:showBib('Borji_etal12eccv')\"\n class=\"bibbase link\">\n \n\t\n\t\n\t\n BibTeX\n <i class=\"fa fa-caret-down\"></i></a>\n \n \n  \n <a class=\"bibbase_abstract_link bibbase link\"\n href=\"javascript:showAbstract('Borji_etal12eccv')\">\n Abstract\n <i class=\"fa fa-caret-down\"></i></a>\n \n \n \n\n \n \n \n</span>\n\n<div class=\"well well-small bibbase\" id=\"bib_Borji_etal12eccv\"\n style=\"display:none\">\n <pre>@inproceedings{ Borji_etal12eccv,\n author = {A. Borji and D. N. Sihite and L. Itti},\n title = {Salient Object Detection: A Benchmark},\n booktitle = {Proc. European Conference on Computer Vision (ECCV), Florence, Italy (LNCS 7573)},\n abstract = {Several salient object detection approaches have been\n published which have been assessed using different\n evaluation scores and datasets resulting in\n discrepancy in model comparison. This calls for a\n methodological framework to compare existing models\n and evaluate their pros and cons. We analyze\n benchmark datasets and scoring techniques and, for\n the first time, provide a quantitative comparison of\n 35 state-of-the-art saliency detection models. We\n find that some models perform consistently better\n than the others. Saliency models that intend to\n predict eye fixations perform lower on segmentation\n datasets compared to salient object detection\n algorithms. Further, we propose combined models\n which show that integration of the few best models\n outperforms all models over other datasets. By\n analyzing the consistency among the best models and\n among humans for each scene, we identify the scenes\n where models or humans fail to detect the most\n salient object. We highlight the current issues and\n propose future research directions.},\n month = {Oct},\n year = {2012},\n pages = {414-429},\n review = {full/conf},\n type = {bu;td;mod;cv},\n if = {2012 acceptance rate: 25.0%},\n file = {http://ilab.usc.edu/publications/doc/Borji_etal12eccv.pdf}\n}</pre>\n</div>\n\n\n<div class=\"well well-small bibbase\" id=\"abstract_Borji_etal12eccv\"\n style=\"display:none\">\n Several salient object detection approaches have been published which have been assessed using different evaluation scores and datasets resulting in discrepancy in model comparison. This calls for a methodological framework to compare existing models and evaluate their pros and cons. We analyze benchmark datasets and scoring techniques and, for the first time, provide a quantitative comparison of 35 state-of-the-art saliency detection models. We find that some models perform consistently better than the others. Saliency models that intend to predict eye fixations perform lower on segmentation datasets compared to salient object detection algorithms. Further, we propose combined models which show that integration of the few best models outperforms all models over other datasets. By analyzing the consistency among the best models and among humans for each scene, we identify the scenes where models or humans fail to detect the most salient object. We highlight the current issues and propose future research directions.\n</div>\n\n\n</div>\n","downloads":0,"bibbaseid":"borji-sihite-itti-salientobjectdetectionabenchmark-2012","role":"author","year":"2012","type":"bu;td;mod;cv","title":"Salient Object Detection: A Benchmark","review":"full/conf","pages":"414-429","month":"Oct","key":"Borji_etal12eccv","if":"2012 acceptance rate: 25.0%","id":"Borji_etal12eccv","file":"http://ilab.usc.edu/publications/doc/Borji_etal12eccv.pdf","booktitle":"Proc. European Conference on Computer Vision (ECCV), Florence, Italy (LNCS 7573)","bibtype":"inproceedings","bibtex":"@inproceedings{ Borji_etal12eccv,\n author = {A. Borji and D. N. Sihite and L. Itti},\n title = {Salient Object Detection: A Benchmark},\n booktitle = {Proc. European Conference on Computer Vision (ECCV), Florence, Italy (LNCS 7573)},\n abstract = {Several salient object detection approaches have been\n published which have been assessed using different\n evaluation scores and datasets resulting in\n discrepancy in model comparison. This calls for a\n methodological framework to compare existing models\n and evaluate their pros and cons. We analyze\n benchmark datasets and scoring techniques and, for\n the first time, provide a quantitative comparison of\n 35 state-of-the-art saliency detection models. We\n find that some models perform consistently better\n than the others. Saliency models that intend to\n predict eye fixations perform lower on segmentation\n datasets compared to salient object detection\n algorithms. Further, we propose combined models\n which show that integration of the few best models\n outperforms all models over other datasets. By\n analyzing the consistency among the best models and\n among humans for each scene, we identify the scenes\n where models or humans fail to detect the most\n salient object. We highlight the current issues and\n propose future research directions.},\n month = {Oct},\n year = {2012},\n pages = {414-429},\n review = {full/conf},\n type = {bu;td;mod;cv},\n if = {2012 acceptance rate: 25.0%},\n file = {http://ilab.usc.edu/publications/doc/Borji_etal12eccv.pdf}\n}","author_short":["Borji, A.","Sihite, D.<nbsp>N.","Itti, L."],"author":["Borji, A.","Sihite, D. N.","Itti, L."],"abstract":"Several salient object detection approaches have been published which have been assessed using different evaluation scores and datasets resulting in discrepancy in model comparison. This calls for a methodological framework to compare existing models and evaluate their pros and cons. We analyze benchmark datasets and scoring techniques and, for the first time, provide a quantitative comparison of 35 state-of-the-art saliency detection models. We find that some models perform consistently better than the others. Saliency models that intend to predict eye fixations perform lower on segmentation datasets compared to salient object detection algorithms. Further, we propose combined models which show that integration of the few best models outperforms all models over other datasets. By analyzing the consistency among the best models and among humans for each scene, we identify the scenes where models or humans fail to detect the most salient object. We highlight the current issues and propose future research directions."},"bibtype":"inproceedings","biburl":"http://ilab.usc.edu/publications/src/ilab.bib","downloads":0,"search_terms":["salient","object","detection","benchmark","borji","sihite","itti"],"title":"Salient Object Detection: A Benchmark","year":2012,"dataSources":["wedBDxEpNXNCLZ2sZ"]}