Robust frontal view search using multi-camera constrained Isomap

Chao Wang, Xubo Song

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Many 2D face processing algorithms can perform better using frontal or near frontal face. In this paper, we present a robust frontal view search method, and apply it to a scene where a rotating head is simultaneously captured by multiple cameras. Our method is based on manifold embedding using Isomap, with the assumption that with head pose being the only variable, the face images should lie in a smooth and low dimensional manifold. We constrain Isomap by (1) a spatio-temporal constraint which improves the neighborhood graph of Isomap to make the manifold retain smooth and low-dimensional under uncontrolled conditions; (2) a multi-camera constraint which reflects the relative pose of multiple simultaneous video streams to adjust the graph distance of manifold in iteration. Using the constrained Isomap, the frontal view is in the vertex of the parabola-shaped manifold in 2D embedding. The experiments show that the results have high reliability.

Original languageEnglish (US)
Title of host publicationProceedings - International Conference on Image Processing, ICIP
Pages1017-1020
Number of pages4
DOIs
StatePublished - 2012
Event2012 19th IEEE International Conference on Image Processing, ICIP 2012 - Lake Buena Vista, FL, United States
Duration: Sep 30 2012Oct 3 2012

Other

Other2012 19th IEEE International Conference on Image Processing, ICIP 2012
CountryUnited States
CityLake Buena Vista, FL
Period9/30/1210/3/12

Fingerprint

Cameras
Processing
Experiments

Keywords

  • Frontal view search
  • Isomap
  • manifold learning
  • multi-camera
  • spatio-temporal

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Information Systems

Cite this

Wang, C., & Song, X. (2012). Robust frontal view search using multi-camera constrained Isomap. In Proceedings - International Conference on Image Processing, ICIP (pp. 1017-1020). [6467035] https://doi.org/10.1109/ICIP.2012.6467035

Robust frontal view search using multi-camera constrained Isomap. / Wang, Chao; Song, Xubo.

Proceedings - International Conference on Image Processing, ICIP. 2012. p. 1017-1020 6467035.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Wang, C & Song, X 2012, Robust frontal view search using multi-camera constrained Isomap. in Proceedings - International Conference on Image Processing, ICIP., 6467035, pp. 1017-1020, 2012 19th IEEE International Conference on Image Processing, ICIP 2012, Lake Buena Vista, FL, United States, 9/30/12. https://doi.org/10.1109/ICIP.2012.6467035
Wang C, Song X. Robust frontal view search using multi-camera constrained Isomap. In Proceedings - International Conference on Image Processing, ICIP. 2012. p. 1017-1020. 6467035 https://doi.org/10.1109/ICIP.2012.6467035
Wang, Chao ; Song, Xubo. / Robust frontal view search using multi-camera constrained Isomap. Proceedings - International Conference on Image Processing, ICIP. 2012. pp. 1017-1020
@inproceedings{21cad7c442e44bb2828559537315c043,
title = "Robust frontal view search using multi-camera constrained Isomap",
abstract = "Many 2D face processing algorithms can perform better using frontal or near frontal face. In this paper, we present a robust frontal view search method, and apply it to a scene where a rotating head is simultaneously captured by multiple cameras. Our method is based on manifold embedding using Isomap, with the assumption that with head pose being the only variable, the face images should lie in a smooth and low dimensional manifold. We constrain Isomap by (1) a spatio-temporal constraint which improves the neighborhood graph of Isomap to make the manifold retain smooth and low-dimensional under uncontrolled conditions; (2) a multi-camera constraint which reflects the relative pose of multiple simultaneous video streams to adjust the graph distance of manifold in iteration. Using the constrained Isomap, the frontal view is in the vertex of the parabola-shaped manifold in 2D embedding. The experiments show that the results have high reliability.",
keywords = "Frontal view search, Isomap, manifold learning, multi-camera, spatio-temporal",
author = "Chao Wang and Xubo Song",
year = "2012",
doi = "10.1109/ICIP.2012.6467035",
language = "English (US)",
isbn = "9781467325332",
pages = "1017--1020",
booktitle = "Proceedings - International Conference on Image Processing, ICIP",

}

TY - GEN

T1 - Robust frontal view search using multi-camera constrained Isomap

AU - Wang, Chao

AU - Song, Xubo

PY - 2012

Y1 - 2012

N2 - Many 2D face processing algorithms can perform better using frontal or near frontal face. In this paper, we present a robust frontal view search method, and apply it to a scene where a rotating head is simultaneously captured by multiple cameras. Our method is based on manifold embedding using Isomap, with the assumption that with head pose being the only variable, the face images should lie in a smooth and low dimensional manifold. We constrain Isomap by (1) a spatio-temporal constraint which improves the neighborhood graph of Isomap to make the manifold retain smooth and low-dimensional under uncontrolled conditions; (2) a multi-camera constraint which reflects the relative pose of multiple simultaneous video streams to adjust the graph distance of manifold in iteration. Using the constrained Isomap, the frontal view is in the vertex of the parabola-shaped manifold in 2D embedding. The experiments show that the results have high reliability.

AB - Many 2D face processing algorithms can perform better using frontal or near frontal face. In this paper, we present a robust frontal view search method, and apply it to a scene where a rotating head is simultaneously captured by multiple cameras. Our method is based on manifold embedding using Isomap, with the assumption that with head pose being the only variable, the face images should lie in a smooth and low dimensional manifold. We constrain Isomap by (1) a spatio-temporal constraint which improves the neighborhood graph of Isomap to make the manifold retain smooth and low-dimensional under uncontrolled conditions; (2) a multi-camera constraint which reflects the relative pose of multiple simultaneous video streams to adjust the graph distance of manifold in iteration. Using the constrained Isomap, the frontal view is in the vertex of the parabola-shaped manifold in 2D embedding. The experiments show that the results have high reliability.

KW - Frontal view search

KW - Isomap

KW - manifold learning

KW - multi-camera

KW - spatio-temporal

UR - http://www.scopus.com/inward/record.url?scp=84875819616&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84875819616&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2012.6467035

DO - 10.1109/ICIP.2012.6467035

M3 - Conference contribution

AN - SCOPUS:84875819616

SN - 9781467325332

SP - 1017

EP - 1020

BT - Proceedings - International Conference on Image Processing, ICIP

ER -