Automatic Attention-Based Prioritization of Unconstrained Video for Compression. Itti, L. In Rogowitz, B. & Pappas, T. N., editors, Proc. SPIE Human Vision and Electronic Imaging IX (HVEI04), San Jose, CA, volume 5292, pages 272-283, Bellingham, WA, Jan, 2004. SPIE Press. abstract bibtex We apply a biologically-motivated algorithm that selects visually-salient regions of interest in video streams to multiply-foveated video compression. Regions of high encoding priority are selected based on nonlinear integration of low-level visual cues, mimicking processing in primate occipital and posterior parietal cortex. A dynamic foveation filter then blurs (foveates) every frame, increasingly with distance from high-priority regions. Sixty-three variants of the algorithm with different parameter settings are evaluated against an outdoor video scene, using MPEG-1 and MPEG-4, yielding compression radios of 1.1 to 8.5. Two variants (one with continuously-variable blur proportional to saliency at every pixel, and the other with blur proportional to distance from three independent foveation centers) are validated against eye fixations from 4-6 human observers on 50 video clips (synthetic stimuli, video games, outdoors day and night home video, television newscast, sports, talk-shows, etc). Significant overlap is found between human and algorithmic foveations on every clip with one variant, and on 48 out of 50 clips with the other. Substantial compressed file size reductions by a factor 0.5 on average are obtained for foveated compared to unfoveated clips. These results suggest a general-purpose usefulness of the algorithm in improving compression ratios of unconstrained video.
@inproceedings{ Itti04hvei,
author = { L. Itti },
title = { Automatic Attention-Based Prioritization of Unconstrained
Video for Compression},
year = {2004},
month = {Jan},
abstract = {We apply a biologically-motivated algorithm that selects
visually-salient regions of interest in video streams to
multiply-foveated video compression. Regions of high encoding
priority are selected based on nonlinear integration of low-level
visual cues, mimicking processing in primate occipital and posterior
parietal cortex. A dynamic foveation filter then blurs (foveates)
every frame, increasingly with distance from high-priority regions.
Sixty-three variants of the algorithm with different parameter
settings are evaluated against an outdoor video scene, using MPEG-1
and MPEG-4, yielding compression radios of 1.1 to 8.5. Two variants
(one with continuously-variable blur proportional to saliency at every
pixel, and the other with blur proportional to distance from three
independent foveation centers) are validated against eye fixations
from 4-6 human observers on 50 video clips (synthetic stimuli, video
games, outdoors day and night home video, television newscast, sports,
talk-shows, etc). Significant overlap is found between human and
algorithmic foveations on every clip with one variant, and on 48 out
of 50 clips with the other. Substantial compressed file size
reductions by a factor 0.5 on average are obtained for foveated
compared to unfoveated clips. These results suggest a general-purpose
usefulness of the algorithm in improving compression ratios of
unconstrained video.},
booktitle = { Proc. SPIE Human Vision and Electronic Imaging IX
(HVEI04), San Jose, CA },
volume = {5292},
pages = {272-283},
editor = {B. Rogowitz and T. N. Pappas},
type = { mod;bu;cv },
publisher = {SPIE Press},
address = {Bellingham, WA},
file = { http://iLab.usc.edu/publications/doc/Itti04hvei.pdf },
review = {abs/conf}
}
Downloads: 0
{"_id":{"_str":"5298a19f9eb585cc260007f7"},"__v":0,"authorIDs":[],"author_short":["Itti, L."],"bibbaseid":"itti-automaticattentionbasedprioritizationofunconstrainedvideoforcompression-2004","bibdata":{"html":"<div class=\"bibbase_paper\"> \n\n\n<span class=\"bibbase_paper_titleauthoryear\">\n\t<span class=\"bibbase_paper_title\"><a name=\"Itti04hvei\"> </a>Automatic Attention-Based Prioritization of Unconstrained Video for Compression.</span>\n\t<span class=\"bibbase_paper_author\">\nItti, L.</span>\n\t<!-- <span class=\"bibbase_paper_year\">2004</span>. -->\n</span>\n\n\n\nIn\nRogowitz, B.; and Pappas, T. N., editor, <i>Proc. SPIE Human Vision and Electronic Imaging IX (HVEI04), San Jose, CA</i>, volume 5292, page 272-283, Bellingham, WA, Jan 2004.\n\n\nSPIE Press.\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('Itti04hvei')\"\n class=\"bibbase link\">\n <!-- <img src=\"http://www.bibbase.org/img/filetypes/bib.png\" -->\n\t<!-- alt=\"Automatic Attention-Based Prioritization of Unconstrained Video for Compression [bib]\" -->\n\t<!-- class=\"bibbase_icon\" -->\n\t<!-- style=\"width: 24px; height: 24px; border: 0px; vertical-align: text-top\"><span class=\"bibbase_icon_text\">Bibtex</span> -->\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('Itti04hvei')\">\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_Itti04hvei\"\n style=\"display:none\">\n <pre>@inproceedings{ Itti04hvei,\n author = { L. Itti },\n title = { Automatic Attention-Based Prioritization of Unconstrained \nVideo for Compression},\n year = {2004},\n month = {Jan},\n abstract = {We apply a biologically-motivated algorithm that selects\nvisually-salient regions of interest in video streams to\nmultiply-foveated video compression. Regions of high encoding\npriority are selected based on nonlinear integration of low-level\nvisual cues, mimicking processing in primate occipital and posterior\nparietal cortex. A dynamic foveation filter then blurs (foveates)\nevery frame, increasingly with distance from high-priority regions.\nSixty-three variants of the algorithm with different parameter\nsettings are evaluated against an outdoor video scene, using MPEG-1\nand MPEG-4, yielding compression radios of 1.1 to 8.5. Two variants\n(one with continuously-variable blur proportional to saliency at every\npixel, and the other with blur proportional to distance from three\nindependent foveation centers) are validated against eye fixations\nfrom 4-6 human observers on 50 video clips (synthetic stimuli, video\ngames, outdoors day and night home video, television newscast, sports,\ntalk-shows, etc). Significant overlap is found between human and\nalgorithmic foveations on every clip with one variant, and on 48 out\nof 50 clips with the other. Substantial compressed file size\nreductions by a factor 0.5 on average are obtained for foveated\ncompared to unfoveated clips. These results suggest a general-purpose\nusefulness of the algorithm in improving compression ratios of\nunconstrained video.},\n booktitle = { Proc. SPIE Human Vision and Electronic Imaging IX\n(HVEI04), San Jose, CA },\n volume = {5292},\n pages = {272-283},\n editor = {B. Rogowitz and T. N. Pappas},\n type = { mod;bu;cv },\n publisher = {SPIE Press},\n address = {Bellingham, WA},\n file = { http://iLab.usc.edu/publications/doc/Itti04hvei.pdf },\n review = {abs/conf}\n}</pre>\n</div>\n\n\n<div class=\"well well-small bibbase\" id=\"abstract_Itti04hvei\"\n style=\"display:none\">\n We apply a biologically-motivated algorithm that selects visually-salient regions of interest in video streams to multiply-foveated video compression. Regions of high encoding priority are selected based on nonlinear integration of low-level visual cues, mimicking processing in primate occipital and posterior parietal cortex. A dynamic foveation filter then blurs (foveates) every frame, increasingly with distance from high-priority regions. Sixty-three variants of the algorithm with different parameter settings are evaluated against an outdoor video scene, using MPEG-1 and MPEG-4, yielding compression radios of 1.1 to 8.5. Two variants (one with continuously-variable blur proportional to saliency at every pixel, and the other with blur proportional to distance from three independent foveation centers) are validated against eye fixations from 4-6 human observers on 50 video clips (synthetic stimuli, video games, outdoors day and night home video, television newscast, sports, talk-shows, etc). Significant overlap is found between human and algorithmic foveations on every clip with one variant, and on 48 out of 50 clips with the other. Substantial compressed file size reductions by a factor 0.5 on average are obtained for foveated compared to unfoveated clips. These results suggest a general-purpose usefulness of the algorithm in improving compression ratios of unconstrained video.\n</div>\n\n\n</div>\n","downloads":0,"bibbaseid":"itti-automaticattentionbasedprioritizationofunconstrainedvideoforcompression-2004","role":"author","year":"2004","volume":"5292","type":"mod;bu;cv","title":"Automatic Attention-Based Prioritization of Unconstrained Video for Compression","review":"abs/conf","publisher":"SPIE Press","pages":"272-283","month":"Jan","key":"Itti04hvei","id":"Itti04hvei","file":"http://iLab.usc.edu/publications/doc/Itti04hvei.pdf","editor_short":["Rogowitz, B.","Pappas, T.<nbsp>N."],"editor":["Rogowitz, B.","Pappas, T. N."],"booktitle":"Proc. SPIE Human Vision and Electronic Imaging IX (HVEI04), San Jose, CA","bibtype":"inproceedings","bibtex":"@inproceedings{ Itti04hvei,\n author = { L. Itti },\n title = { Automatic Attention-Based Prioritization of Unconstrained \nVideo for Compression},\n year = {2004},\n month = {Jan},\n abstract = {We apply a biologically-motivated algorithm that selects\nvisually-salient regions of interest in video streams to\nmultiply-foveated video compression. Regions of high encoding\npriority are selected based on nonlinear integration of low-level\nvisual cues, mimicking processing in primate occipital and posterior\nparietal cortex. A dynamic foveation filter then blurs (foveates)\nevery frame, increasingly with distance from high-priority regions.\nSixty-three variants of the algorithm with different parameter\nsettings are evaluated against an outdoor video scene, using MPEG-1\nand MPEG-4, yielding compression radios of 1.1 to 8.5. Two variants\n(one with continuously-variable blur proportional to saliency at every\npixel, and the other with blur proportional to distance from three\nindependent foveation centers) are validated against eye fixations\nfrom 4-6 human observers on 50 video clips (synthetic stimuli, video\ngames, outdoors day and night home video, television newscast, sports,\ntalk-shows, etc). Significant overlap is found between human and\nalgorithmic foveations on every clip with one variant, and on 48 out\nof 50 clips with the other. Substantial compressed file size\nreductions by a factor 0.5 on average are obtained for foveated\ncompared to unfoveated clips. These results suggest a general-purpose\nusefulness of the algorithm in improving compression ratios of\nunconstrained video.},\n booktitle = { Proc. SPIE Human Vision and Electronic Imaging IX\n(HVEI04), San Jose, CA },\n volume = {5292},\n pages = {272-283},\n editor = {B. Rogowitz and T. N. Pappas},\n type = { mod;bu;cv },\n publisher = {SPIE Press},\n address = {Bellingham, WA},\n file = { http://iLab.usc.edu/publications/doc/Itti04hvei.pdf },\n review = {abs/conf}\n}","author_short":["Itti, L."],"author":["Itti, L."],"address":"Bellingham, WA","abstract":"We apply a biologically-motivated algorithm that selects visually-salient regions of interest in video streams to multiply-foveated video compression. Regions of high encoding priority are selected based on nonlinear integration of low-level visual cues, mimicking processing in primate occipital and posterior parietal cortex. A dynamic foveation filter then blurs (foveates) every frame, increasingly with distance from high-priority regions. Sixty-three variants of the algorithm with different parameter settings are evaluated against an outdoor video scene, using MPEG-1 and MPEG-4, yielding compression radios of 1.1 to 8.5. Two variants (one with continuously-variable blur proportional to saliency at every pixel, and the other with blur proportional to distance from three independent foveation centers) are validated against eye fixations from 4-6 human observers on 50 video clips (synthetic stimuli, video games, outdoors day and night home video, television newscast, sports, talk-shows, etc). Significant overlap is found between human and algorithmic foveations on every clip with one variant, and on 48 out of 50 clips with the other. Substantial compressed file size reductions by a factor 0.5 on average are obtained for foveated compared to unfoveated clips. These results suggest a general-purpose usefulness of the algorithm in improving compression ratios of unconstrained video."},"bibtype":"inproceedings","biburl":"http://ilab.usc.edu/publications/src/ilab.bib","downloads":0,"search_terms":["automatic","attention","based","prioritization","unconstrained","video","compression","itti"],"title":"Automatic Attention-Based Prioritization of Unconstrained Video for Compression","year":2004,"dataSources":["wedBDxEpNXNCLZ2sZ"]}