Real-Time Acquisition of High Quality Face Sequences from an Active Pan-Tilt-Zoom Camera

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

19 Citations (Scopus)
506 Downloads (Pure)

Abstract

Traditional still camera-based facial image acquisition systems in surveillance applications produce low quality face images. This is mainly due to the distance between the camera and subjects of interest. Furthermore, people in such videos usually move around, change their head poses, and facial expressions. Moreover, the imaging conditions like illumination, occlusion, and noise may change. These all aggregate the quality of most of the detected face images in terms of measures like resolution, pose, brightness, and sharpness. To deal with these problems this paper presents an active camera-based real-time high-quality face image acquisition system, which utilizes pan-tilt-zoom parameters of a camera to focus on a human face in a scene and employs a face quality assessment method to log the best quality faces from the captured frames. The system consists of four modules: face detection, camera control, face tracking, and face quality assessment before logging. Experimental results show that the proposed system can effectively log the high quality faces from the active camera in real-time (an average of 61.74ms was spent per frame) with an accuracy of 85.27% compared to human annotated data.

Original languageEnglish
Title of host publication2013 10th IEEE International Conference on Advanced Video and Signal-Based Surveillance (AVSS) : Workshop on Low-Resolution Face Analysis (LRFA 2013)
Number of pages6
PublisherIEEE Press
Publication date27 Aug 2013
Pages443-448
ISBN (Print)978-1-4799-0703-8, 9781479907045
ISBN (Electronic)9781479907021
DOIs
Publication statusPublished - 27 Aug 2013
EventAVSS 2013 - Kraków, Poland
Duration: 27 Aug 201330 Aug 2013

Conference

ConferenceAVSS 2013
CountryPoland
CityKraków
Period27/08/201330/08/2013

Fingerprint

Cameras
Image acquisition
Face recognition
Luminance
Lighting
Imaging techniques

Keywords

  • Face-log
  • Active PTZ Camera
  • Face Quality Assessment
  • Face detection
  • Face Tracking

Cite this

Haque, M. A., Nasrollahi, K., & Moeslund, T. B. (2013). Real-Time Acquisition of High Quality Face Sequences from an Active Pan-Tilt-Zoom Camera. In 2013 10th IEEE International Conference on Advanced Video and Signal-Based Surveillance (AVSS): Workshop on Low-Resolution Face Analysis (LRFA 2013) (pp. 443-448). IEEE Press. https://doi.org/10.1109/AVSS.2013.6636680
Haque, Mohammad A. ; Nasrollahi, Kamal ; Moeslund, Thomas B. / Real-Time Acquisition of High Quality Face Sequences from an Active Pan-Tilt-Zoom Camera. 2013 10th IEEE International Conference on Advanced Video and Signal-Based Surveillance (AVSS): Workshop on Low-Resolution Face Analysis (LRFA 2013). IEEE Press, 2013. pp. 443-448
@inproceedings{f6069b601f5d4fb9b25dd78888e39377,
title = "Real-Time Acquisition of High Quality Face Sequences from an Active Pan-Tilt-Zoom Camera",
abstract = "Traditional still camera-based facial image acquisition systems in surveillance applications produce low quality face images. This is mainly due to the distance between the camera and subjects of interest. Furthermore, people in such videos usually move around, change their head poses, and facial expressions. Moreover, the imaging conditions like illumination, occlusion, and noise may change. These all aggregate the quality of most of the detected face images in terms of measures like resolution, pose, brightness, and sharpness. To deal with these problems this paper presents an active camera-based real-time high-quality face image acquisition system, which utilizes pan-tilt-zoom parameters of a camera to focus on a human face in a scene and employs a face quality assessment method to log the best quality faces from the captured frames. The system consists of four modules: face detection, camera control, face tracking, and face quality assessment before logging. Experimental results show that the proposed system can effectively log the high quality faces from the active camera in real-time (an average of 61.74ms was spent per frame) with an accuracy of 85.27{\%} compared to human annotated data.",
keywords = "Face-log, Active PTZ Camera, Face Quality Assessment, Face detection, Face Tracking",
author = "Haque, {Mohammad A.} and Kamal Nasrollahi and Moeslund, {Thomas B.}",
year = "2013",
month = "8",
day = "27",
doi = "10.1109/AVSS.2013.6636680",
language = "English",
isbn = "978-1-4799-0703-8",
pages = "443--448",
booktitle = "2013 10th IEEE International Conference on Advanced Video and Signal-Based Surveillance (AVSS)",
publisher = "IEEE Press",

}

Haque, MA, Nasrollahi, K & Moeslund, TB 2013, Real-Time Acquisition of High Quality Face Sequences from an Active Pan-Tilt-Zoom Camera. in 2013 10th IEEE International Conference on Advanced Video and Signal-Based Surveillance (AVSS): Workshop on Low-Resolution Face Analysis (LRFA 2013). IEEE Press, pp. 443-448, AVSS 2013, Kraków, Poland, 27/08/2013. https://doi.org/10.1109/AVSS.2013.6636680

Real-Time Acquisition of High Quality Face Sequences from an Active Pan-Tilt-Zoom Camera. / Haque, Mohammad A.; Nasrollahi, Kamal; Moeslund, Thomas B.

2013 10th IEEE International Conference on Advanced Video and Signal-Based Surveillance (AVSS): Workshop on Low-Resolution Face Analysis (LRFA 2013). IEEE Press, 2013. p. 443-448.

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

TY - GEN

T1 - Real-Time Acquisition of High Quality Face Sequences from an Active Pan-Tilt-Zoom Camera

AU - Haque, Mohammad A.

AU - Nasrollahi, Kamal

AU - Moeslund, Thomas B.

PY - 2013/8/27

Y1 - 2013/8/27

N2 - Traditional still camera-based facial image acquisition systems in surveillance applications produce low quality face images. This is mainly due to the distance between the camera and subjects of interest. Furthermore, people in such videos usually move around, change their head poses, and facial expressions. Moreover, the imaging conditions like illumination, occlusion, and noise may change. These all aggregate the quality of most of the detected face images in terms of measures like resolution, pose, brightness, and sharpness. To deal with these problems this paper presents an active camera-based real-time high-quality face image acquisition system, which utilizes pan-tilt-zoom parameters of a camera to focus on a human face in a scene and employs a face quality assessment method to log the best quality faces from the captured frames. The system consists of four modules: face detection, camera control, face tracking, and face quality assessment before logging. Experimental results show that the proposed system can effectively log the high quality faces from the active camera in real-time (an average of 61.74ms was spent per frame) with an accuracy of 85.27% compared to human annotated data.

AB - Traditional still camera-based facial image acquisition systems in surveillance applications produce low quality face images. This is mainly due to the distance between the camera and subjects of interest. Furthermore, people in such videos usually move around, change their head poses, and facial expressions. Moreover, the imaging conditions like illumination, occlusion, and noise may change. These all aggregate the quality of most of the detected face images in terms of measures like resolution, pose, brightness, and sharpness. To deal with these problems this paper presents an active camera-based real-time high-quality face image acquisition system, which utilizes pan-tilt-zoom parameters of a camera to focus on a human face in a scene and employs a face quality assessment method to log the best quality faces from the captured frames. The system consists of four modules: face detection, camera control, face tracking, and face quality assessment before logging. Experimental results show that the proposed system can effectively log the high quality faces from the active camera in real-time (an average of 61.74ms was spent per frame) with an accuracy of 85.27% compared to human annotated data.

KW - Face-log

KW - Active PTZ Camera

KW - Face Quality Assessment

KW - Face detection

KW - Face Tracking

U2 - 10.1109/AVSS.2013.6636680

DO - 10.1109/AVSS.2013.6636680

M3 - Article in proceeding

SN - 978-1-4799-0703-8

SN - 9781479907045

SP - 443

EP - 448

BT - 2013 10th IEEE International Conference on Advanced Video and Signal-Based Surveillance (AVSS)

PB - IEEE Press

ER -

Haque MA, Nasrollahi K, Moeslund TB. Real-Time Acquisition of High Quality Face Sequences from an Active Pan-Tilt-Zoom Camera. In 2013 10th IEEE International Conference on Advanced Video and Signal-Based Surveillance (AVSS): Workshop on Low-Resolution Face Analysis (LRFA 2013). IEEE Press. 2013. p. 443-448 https://doi.org/10.1109/AVSS.2013.6636680