iFR: interactively pose corrected face recognition

Simon Nash, Mark Rhodes, Joanna Isabelle Olszewska

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

Abstract

Although face recognition applications are growing, robust face recognition is still a challenging task due e.g. to variations on face poses, facial expressions, or lighting conditions. In this paper, we propose a new method which allows both automatic face detection and recognition and incorporates an interactive selection of facial features in conjunction with our new pose-correction algorithm. Our resulting system we call iFR successfully recognizes faces across pose, while being computationally efficient and outperforming standard approaches, as demonstrated in tests carried out on publicly available standard datasets.
Original languageEnglish
Title of host publicationProceedings of the 9th International Joint Conference on Biomedical Engineering Systems and Technologies
Subtitle of host publicationBIOSIGNALS, (BIOSTEC 2016)
Pages106-112
Number of pages7
Volume4
DOIs
Publication statusPublished - 2016
Externally publishedYes

Fingerprint

Face recognition
Lighting

Keywords

  • Face Detection
  • Face Recognition
  • AdaBoost
  • Template
  • Facial Features
  • Eigenfaces
  • Pose Correction
  • Human-Computer Interactive Systems

Cite this

Nash, S., Rhodes, M., & Olszewska, J. I. (2016). iFR: interactively pose corrected face recognition. In Proceedings of the 9th International Joint Conference on Biomedical Engineering Systems and Technologies: BIOSIGNALS, (BIOSTEC 2016) (Vol. 4, pp. 106-112) https://doi.org/10.5220/0005857801060112
Nash, Simon ; Rhodes, Mark ; Olszewska, Joanna Isabelle. / iFR : interactively pose corrected face recognition. Proceedings of the 9th International Joint Conference on Biomedical Engineering Systems and Technologies: BIOSIGNALS, (BIOSTEC 2016). Vol. 4 2016. pp. 106-112
@inproceedings{e0c0e06bfe7e450a91bb02f8f947da0e,
title = "iFR: interactively pose corrected face recognition",
abstract = "Although face recognition applications are growing, robust face recognition is still a challenging task due e.g. to variations on face poses, facial expressions, or lighting conditions. In this paper, we propose a new method which allows both automatic face detection and recognition and incorporates an interactive selection of facial features in conjunction with our new pose-correction algorithm. Our resulting system we call iFR successfully recognizes faces across pose, while being computationally efficient and outperforming standard approaches, as demonstrated in tests carried out on publicly available standard datasets.",
keywords = "Face Detection, Face Recognition, AdaBoost, Template, Facial Features, Eigenfaces, Pose Correction, Human-Computer Interactive Systems",
author = "Simon Nash and Mark Rhodes and Olszewska, {Joanna Isabelle}",
year = "2016",
doi = "10.5220/0005857801060112",
language = "English",
isbn = "9789897581700",
volume = "4",
pages = "106--112",
booktitle = "Proceedings of the 9th International Joint Conference on Biomedical Engineering Systems and Technologies",

}

Nash, S, Rhodes, M & Olszewska, JI 2016, iFR: interactively pose corrected face recognition. in Proceedings of the 9th International Joint Conference on Biomedical Engineering Systems and Technologies: BIOSIGNALS, (BIOSTEC 2016). vol. 4, pp. 106-112. https://doi.org/10.5220/0005857801060112

iFR : interactively pose corrected face recognition. / Nash, Simon; Rhodes, Mark; Olszewska, Joanna Isabelle.

Proceedings of the 9th International Joint Conference on Biomedical Engineering Systems and Technologies: BIOSIGNALS, (BIOSTEC 2016). Vol. 4 2016. p. 106-112.

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

TY - GEN

T1 - iFR

T2 - interactively pose corrected face recognition

AU - Nash, Simon

AU - Rhodes, Mark

AU - Olszewska, Joanna Isabelle

PY - 2016

Y1 - 2016

N2 - Although face recognition applications are growing, robust face recognition is still a challenging task due e.g. to variations on face poses, facial expressions, or lighting conditions. In this paper, we propose a new method which allows both automatic face detection and recognition and incorporates an interactive selection of facial features in conjunction with our new pose-correction algorithm. Our resulting system we call iFR successfully recognizes faces across pose, while being computationally efficient and outperforming standard approaches, as demonstrated in tests carried out on publicly available standard datasets.

AB - Although face recognition applications are growing, robust face recognition is still a challenging task due e.g. to variations on face poses, facial expressions, or lighting conditions. In this paper, we propose a new method which allows both automatic face detection and recognition and incorporates an interactive selection of facial features in conjunction with our new pose-correction algorithm. Our resulting system we call iFR successfully recognizes faces across pose, while being computationally efficient and outperforming standard approaches, as demonstrated in tests carried out on publicly available standard datasets.

KW - Face Detection

KW - Face Recognition

KW - AdaBoost

KW - Template

KW - Facial Features

KW - Eigenfaces

KW - Pose Correction

KW - Human-Computer Interactive Systems

U2 - 10.5220/0005857801060112

DO - 10.5220/0005857801060112

M3 - Conference contribution

SN - 9789897581700

VL - 4

SP - 106

EP - 112

BT - Proceedings of the 9th International Joint Conference on Biomedical Engineering Systems and Technologies

ER -

Nash S, Rhodes M, Olszewska JI. iFR: interactively pose corrected face recognition. In Proceedings of the 9th International Joint Conference on Biomedical Engineering Systems and Technologies: BIOSIGNALS, (BIOSTEC 2016). Vol. 4. 2016. p. 106-112 https://doi.org/10.5220/0005857801060112