Home > Research > Publications & Outputs > Integrated Deep Model for for Face Detection an...

Links

Text available via DOI:

View graph of relations

Integrated Deep Model for for Face Detection and Landmark Localization From "In The Wild " Images

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

Integrated Deep Model for for Face Detection and Landmark Localization From "In The Wild " Images. / Storey, Gary; Bouridane, Ahmed; Jiang, Richard.
In: IEEE Access, Vol. 6, 19.11.2018, p. 74442-74452.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

APA

Vancouver

Storey G, Bouridane A, Jiang R. Integrated Deep Model for for Face Detection and Landmark Localization From "In The Wild " Images. IEEE Access. 2018 Nov 19;6:74442-74452. doi: 10.1109/ACCESS.2018.2882227

Author

Storey, Gary ; Bouridane, Ahmed ; Jiang, Richard. / Integrated Deep Model for for Face Detection and Landmark Localization From "In The Wild " Images. In: IEEE Access. 2018 ; Vol. 6. pp. 74442-74452.

Bibtex

@article{016cc01bad394c208ecac01ab286e476,
title = "Integrated Deep Model for for Face Detection and Landmark Localization From {"}In The Wild {"} Images",
abstract = "The tasks of face detection and landmark localisation are a key foundation for many facial analysis applications, while great advancements have been achieved in recent years there are still challenges to increase the precision of face detection. Within this paper, we present our novel method the IntegratedDeep Model (IDM), fusing two state-of-the-art deep learning architectures, namely, Faster R-CNN and a stacked hourglass for improved face detection precision and accurate landmark localisation. Integration is achieved through the application of a novel optimisation function and is shown in experimental evaluation to increase accuracy of face detection specifically precision by reducing false positive detection{\textquoteright}s by an average of 62%. Our proposed IDM method is evaluated on the Annotated Faces In-The-Wild, Annotated Facial Landmarks In The Wild and the Face Detection Dataset and Benchmark face detection test sets and shows a high level of recall and precision when compared with previously proposed methods. Landmark localisation is evaluated on the Annotated Faces In-The-Wild and 300-W test sets, this specifically focuseson localisation accuracy from detected face bounding boxes when compared with baseline evaluations using ground truth bounding boxes. Our findings highlight only a small 0.005% maximum increase in error which is more profound for the subset of facial landmarks which border the face.",
keywords = "Computer vision, face detection, machine learning",
author = "Gary Storey and Ahmed Bouridane and Richard Jiang",
year = "2018",
month = nov,
day = "19",
doi = "10.1109/ACCESS.2018.2882227",
language = "English",
volume = "6",
pages = "74442--74452",
journal = "IEEE Access",
issn = "2169-3536",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

RIS

TY - JOUR

T1 - Integrated Deep Model for for Face Detection and Landmark Localization From "In The Wild " Images

AU - Storey, Gary

AU - Bouridane, Ahmed

AU - Jiang, Richard

PY - 2018/11/19

Y1 - 2018/11/19

N2 - The tasks of face detection and landmark localisation are a key foundation for many facial analysis applications, while great advancements have been achieved in recent years there are still challenges to increase the precision of face detection. Within this paper, we present our novel method the IntegratedDeep Model (IDM), fusing two state-of-the-art deep learning architectures, namely, Faster R-CNN and a stacked hourglass for improved face detection precision and accurate landmark localisation. Integration is achieved through the application of a novel optimisation function and is shown in experimental evaluation to increase accuracy of face detection specifically precision by reducing false positive detection’s by an average of 62%. Our proposed IDM method is evaluated on the Annotated Faces In-The-Wild, Annotated Facial Landmarks In The Wild and the Face Detection Dataset and Benchmark face detection test sets and shows a high level of recall and precision when compared with previously proposed methods. Landmark localisation is evaluated on the Annotated Faces In-The-Wild and 300-W test sets, this specifically focuseson localisation accuracy from detected face bounding boxes when compared with baseline evaluations using ground truth bounding boxes. Our findings highlight only a small 0.005% maximum increase in error which is more profound for the subset of facial landmarks which border the face.

AB - The tasks of face detection and landmark localisation are a key foundation for many facial analysis applications, while great advancements have been achieved in recent years there are still challenges to increase the precision of face detection. Within this paper, we present our novel method the IntegratedDeep Model (IDM), fusing two state-of-the-art deep learning architectures, namely, Faster R-CNN and a stacked hourglass for improved face detection precision and accurate landmark localisation. Integration is achieved through the application of a novel optimisation function and is shown in experimental evaluation to increase accuracy of face detection specifically precision by reducing false positive detection’s by an average of 62%. Our proposed IDM method is evaluated on the Annotated Faces In-The-Wild, Annotated Facial Landmarks In The Wild and the Face Detection Dataset and Benchmark face detection test sets and shows a high level of recall and precision when compared with previously proposed methods. Landmark localisation is evaluated on the Annotated Faces In-The-Wild and 300-W test sets, this specifically focuseson localisation accuracy from detected face bounding boxes when compared with baseline evaluations using ground truth bounding boxes. Our findings highlight only a small 0.005% maximum increase in error which is more profound for the subset of facial landmarks which border the face.

KW - Computer vision

KW - face detection

KW - machine learning

U2 - 10.1109/ACCESS.2018.2882227

DO - 10.1109/ACCESS.2018.2882227

M3 - Journal article

VL - 6

SP - 74442

EP - 74452

JO - IEEE Access

JF - IEEE Access

SN - 2169-3536

ER -