Biologically-Inspired Face Detection: Non-Brute-Force-Search Approach

Biologically-Inspired Face Detection: Non-Brute-Force-Search Approach. Siagian, C. & Itti, L. In First IEEE-CVPR International Workshop on Face Processing in Video, pages 62-69, Jun, 2004.
abstract bibtex

We present a biologically-inspired face detection system. The system applies notions such as saliency, gist, and gaze to localize a face without performing blind spatial search. The saliency model consists of highly parallel low-level computations that operate in domains such as intensity, orientation, and color. It is used to direct attention to a set of conspicuous locations in an image as starting points. The gist model, computed in parallel with the saliency model, estimates holistic image characteristics such as dominant contours and magnitude in high and low spatial frequency bands. We are limiting its use to predicting the likely head size based on the entire scene. Also, instead of identifying face as a single entity, this system performs detection by parts and uses spatial configuration constraints to be robust against occlusion and perspective.

@inproceedings{ Siagian_Itti04fpiv,
  author = {C. Siagian and L. Itti},
  title = {Biologically-Inspired Face Detection: Non-Brute-Force-Search Approach},
  booktitle = {First IEEE-CVPR International Workshop on Face Processing in Video},
  abstract = {We present a biologically-inspired face detection
system. The system applies notions such as saliency, gist, and gaze to
localize a face without performing blind spatial search. The saliency
model consists of highly parallel low-level computations that operate
in domains such as intensity, orientation, and color. It is used to
direct attention to a set of conspicuous locations in an image as
starting points. The gist model, computed in parallel with the
saliency model, estimates holistic image characteristics such as
dominant contours and magnitude in high and low spatial frequency
bands. We are limiting its use to predicting the likely head size
based on the entire scene. Also, instead of identifying face as a
single entity, this system performs detection by parts and uses
spatial configuration constraints to be robust against occlusion and
perspective.},
  month = {Jun},
  pages = {62-69},
  year = {2004},
  type = {cv ; bu ; sc},
  file = { http://iLab.usc.edu/publications/doc/Siagian_Itti04fpiv.pdf },
  review = {full/wkshp}
}

Downloads: 0

{"_id":{"_str":"5298a1a19eb585cc260008f9"},"__v":0,"authorIDs":[],"author_short":["Siagian, C.","Itti, L."],"bibbaseid":"siagian-itti-biologicallyinspiredfacedetectionnonbruteforcesearchapproach-2004","bibdata":{"html":"<div class=\"bibbase_paper\"> \n\n\n<span class=\"bibbase_paper_titleauthoryear\">\n\t<span class=\"bibbase_paper_title\"><a name=\"Siagian_Itti04fpiv\"> </a>Biologically-Inspired Face Detection: Non-Brute-Force-Search Approach.</span>\n\t<span class=\"bibbase_paper_author\">\nSiagian, C.; and Itti, L.</span>\n\t\n</span>\n\n\n\nIn\n<i>First IEEE-CVPR International Workshop on Face Processing in Video</i>, page 62-69, Jun 2004.\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('Siagian_Itti04fpiv')\"\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('Siagian_Itti04fpiv')\">\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_Siagian_Itti04fpiv\"\n style=\"display:none\">\n <pre>@inproceedings{ Siagian_Itti04fpiv,\n author = {C. Siagian and L. Itti},\n title = {Biologically-Inspired Face Detection: Non-Brute-Force-Search Approach},\n booktitle = {First IEEE-CVPR International Workshop on Face Processing in Video},\n abstract = {We present a biologically-inspired face detection\nsystem. The system applies notions such as saliency, gist, and gaze to\nlocalize a face without performing blind spatial search. The saliency\nmodel consists of highly parallel low-level computations that operate\nin domains such as intensity, orientation, and color. It is used to\ndirect attention to a set of conspicuous locations in an image as\nstarting points. The gist model, computed in parallel with the\nsaliency model, estimates holistic image characteristics such as\ndominant contours and magnitude in high and low spatial frequency\nbands. We are limiting its use to predicting the likely head size\nbased on the entire scene. Also, instead of identifying face as a\nsingle entity, this system performs detection by parts and uses\nspatial configuration constraints to be robust against occlusion and\nperspective.},\n month = {Jun},\n pages = {62-69},\n year = {2004},\n type = {cv ; bu ; sc},\n file = { http://iLab.usc.edu/publications/doc/Siagian_Itti04fpiv.pdf },\n review = {full/wkshp}\n}</pre>\n</div>\n\n\n<div class=\"well well-small bibbase\" id=\"abstract_Siagian_Itti04fpiv\"\n style=\"display:none\">\n We present a biologically-inspired face detection system. The system applies notions such as saliency, gist, and gaze to localize a face without performing blind spatial search. The saliency model consists of highly parallel low-level computations that operate in domains such as intensity, orientation, and color. It is used to direct attention to a set of conspicuous locations in an image as starting points. The gist model, computed in parallel with the saliency model, estimates holistic image characteristics such as dominant contours and magnitude in high and low spatial frequency bands. We are limiting its use to predicting the likely head size based on the entire scene. Also, instead of identifying face as a single entity, this system performs detection by parts and uses spatial configuration constraints to be robust against occlusion and perspective.\n</div>\n\n\n</div>\n","downloads":0,"bibbaseid":"siagian-itti-biologicallyinspiredfacedetectionnonbruteforcesearchapproach-2004","role":"author","year":"2004","type":"cv ; bu ; sc","title":"Biologically-Inspired Face Detection: Non-Brute-Force-Search Approach","review":"full/wkshp","pages":"62-69","month":"Jun","key":"Siagian_Itti04fpiv","id":"Siagian_Itti04fpiv","file":"http://iLab.usc.edu/publications/doc/Siagian_Itti04fpiv.pdf","booktitle":"First IEEE-CVPR International Workshop on Face Processing in Video","bibtype":"inproceedings","bibtex":"@inproceedings{ Siagian_Itti04fpiv,\n author = {C. Siagian and L. Itti},\n title = {Biologically-Inspired Face Detection: Non-Brute-Force-Search Approach},\n booktitle = {First IEEE-CVPR International Workshop on Face Processing in Video},\n abstract = {We present a biologically-inspired face detection\nsystem. The system applies notions such as saliency, gist, and gaze to\nlocalize a face without performing blind spatial search. The saliency\nmodel consists of highly parallel low-level computations that operate\nin domains such as intensity, orientation, and color. It is used to\ndirect attention to a set of conspicuous locations in an image as\nstarting points. The gist model, computed in parallel with the\nsaliency model, estimates holistic image characteristics such as\ndominant contours and magnitude in high and low spatial frequency\nbands. We are limiting its use to predicting the likely head size\nbased on the entire scene. Also, instead of identifying face as a\nsingle entity, this system performs detection by parts and uses\nspatial configuration constraints to be robust against occlusion and\nperspective.},\n month = {Jun},\n pages = {62-69},\n year = {2004},\n type = {cv ; bu ; sc},\n file = { http://iLab.usc.edu/publications/doc/Siagian_Itti04fpiv.pdf },\n review = {full/wkshp}\n}","author_short":["Siagian, C.","Itti, L."],"author":["Siagian, C.","Itti, L."],"abstract":"We present a biologically-inspired face detection system. The system applies notions such as saliency, gist, and gaze to localize a face without performing blind spatial search. The saliency model consists of highly parallel low-level computations that operate in domains such as intensity, orientation, and color. It is used to direct attention to a set of conspicuous locations in an image as starting points. The gist model, computed in parallel with the saliency model, estimates holistic image characteristics such as dominant contours and magnitude in high and low spatial frequency bands. We are limiting its use to predicting the likely head size based on the entire scene. Also, instead of identifying face as a single entity, this system performs detection by parts and uses spatial configuration constraints to be robust against occlusion and perspective."},"bibtype":"inproceedings","biburl":"http://ilab.usc.edu/publications/src/ilab.bib","downloads":0,"search_terms":["biologically","inspired","face","detection","non","brute","force","search","approach","siagian","itti"],"title":"Biologically-Inspired Face Detection: Non-Brute-Force-Search Approach","year":2004,"dataSources":["wedBDxEpNXNCLZ2sZ"]}