Tracking faces

Stephen McKenna; Shaogang Gong

doi:10.1109/AFGR.1996.557276

Tracking faces

Stephen McKenna, Shaogang Gong

Computing

Research output: Contribution to conference › Paper › peer-review

38 Citations (Scopus)

Abstract

Robust tracking and segmentation of faces is a prerequisite for face analysis and recognition. In this paper, we describe an approach to this problem which is well suited to surveillance applications with poorly constrained viewing conditions. It integrates motion-based tracking with model-based face detection to produce segmented face sequences from complex scenes containing several people. The motion of moving image contours was estimated using temporal convolution and a temporally consistent list of moving objects was maintained. Objects were tracked using Kalman filters. Faces were detected using a neural network. The essence of the system is that the motion tracker is able to focus attention for a face detection network whilst the latter is used to aid the tracking process.

Original language	English
Pages	271-276
Number of pages	6
DOIs	https://doi.org/10.1109/AFGR.1996.557276
Publication status	Published - 1 Dec 1996
Event	Proceedings of the 1996 2nd International Conference on Automatic Face and Gesture Recognition - Killington, VT, USA Duration: 14 Oct 1996 → 16 Oct 1996

Conference

Conference	Proceedings of the 1996 2nd International Conference on Automatic Face and Gesture Recognition
City	Killington, VT, USA
Period	14/10/96 → 16/10/96

Keywords

Face detection
Tracking
Robustness
Face recognition
Surveillance
Image segmentation
Layout
Motion estimation
Convolution
Neural networks

ASJC Scopus subject areas

Computer Vision and Pattern Recognition

Access to Document

10.1109/AFGR.1996.557276Licence: No Licence / Unknown

Cite this

@conference{9214ae2b644747c0a413bea745939569,

title = "Tracking faces",

abstract = "Robust tracking and segmentation of faces is a prerequisite for face analysis and recognition. In this paper, we describe an approach to this problem which is well suited to surveillance applications with poorly constrained viewing conditions. It integrates motion-based tracking with model-based face detection to produce segmented face sequences from complex scenes containing several people. The motion of moving image contours was estimated using temporal convolution and a temporally consistent list of moving objects was maintained. Objects were tracked using Kalman filters. Faces were detected using a neural network. The essence of the system is that the motion tracker is able to focus attention for a face detection network whilst the latter is used to aid the tracking process.",

keywords = "Face detection, Tracking, Robustness, Face recognition, Surveillance, Image segmentation, Layout, Motion estimation, Convolution, Neural networks",

author = "Stephen McKenna and Shaogang Gong",

year = "1996",

month = dec,

day = "1",

doi = "10.1109/AFGR.1996.557276",

language = "English",

pages = "271--276",

note = "Proceedings of the 1996 2nd International Conference on Automatic Face and Gesture Recognition ; Conference date: 14-10-1996 Through 16-10-1996",

}

TY - CONF

T1 - Tracking faces

AU - McKenna, Stephen

AU - Gong, Shaogang

PY - 1996/12/1

Y1 - 1996/12/1

N2 - Robust tracking and segmentation of faces is a prerequisite for face analysis and recognition. In this paper, we describe an approach to this problem which is well suited to surveillance applications with poorly constrained viewing conditions. It integrates motion-based tracking with model-based face detection to produce segmented face sequences from complex scenes containing several people. The motion of moving image contours was estimated using temporal convolution and a temporally consistent list of moving objects was maintained. Objects were tracked using Kalman filters. Faces were detected using a neural network. The essence of the system is that the motion tracker is able to focus attention for a face detection network whilst the latter is used to aid the tracking process.

AB - Robust tracking and segmentation of faces is a prerequisite for face analysis and recognition. In this paper, we describe an approach to this problem which is well suited to surveillance applications with poorly constrained viewing conditions. It integrates motion-based tracking with model-based face detection to produce segmented face sequences from complex scenes containing several people. The motion of moving image contours was estimated using temporal convolution and a temporally consistent list of moving objects was maintained. Objects were tracked using Kalman filters. Faces were detected using a neural network. The essence of the system is that the motion tracker is able to focus attention for a face detection network whilst the latter is used to aid the tracking process.

KW - Face detection

KW - Tracking

KW - Robustness

KW - Face recognition

KW - Surveillance

KW - Image segmentation

KW - Layout

KW - Motion estimation

KW - Convolution

KW - Neural networks

UR - http://www.scopus.com/inward/record.url?scp=0030396986&partnerID=8YFLogxK

U2 - 10.1109/AFGR.1996.557276

DO - 10.1109/AFGR.1996.557276

M3 - Paper

AN - SCOPUS:0030396986

SP - 271

EP - 276

T2 - Proceedings of the 1996 2nd International Conference on Automatic Face and Gesture Recognition

Y2 - 14 October 1996 through 16 October 1996

ER -

Tracking faces

Abstract

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this