Multi-Branch Fully Convolutional Network for Face Detection

Handle URI:
http://hdl.handle.net/10754/626520
Title:
Multi-Branch Fully Convolutional Network for Face Detection
Authors:
Bai, Yancheng; Ghanem, Bernard ( 0000-0002-5534-587X )
Abstract:
Face detection is a fundamental problem in computer vision. It is still a challenging task in unconstrained conditions due to significant variations in scale, pose, expressions, and occlusion. In this paper, we propose a multi-branch fully convolutional network (MB-FCN) for face detection, which considers both efficiency and effectiveness in the design process. Our MB-FCN detector can deal with faces at all scale ranges with only a single pass through the backbone network. As such, our MB-FCN model saves computation and thus is more efficient, compared to previous methods that make multiple passes. For each branch, the specific skip connections of the convolutional feature maps at different layers are exploited to represent faces in specific scale ranges. Specifically, small faces can be represented with both shallow fine-grained and deep powerful coarse features. With this representation, superior improvement in performance is registered for the task of detecting small faces. We test our MB-FCN detector on two public face detection benchmarks, including FDDB and WIDER FACE. Extensive experiments show that our detector outperforms state-of-the-art methods on all these datasets in general and by a substantial margin on the most challenging among them (e.g. WIDER FACE Hard subset). Also, MB-FCN runs at 15 FPS on a GPU for images of size 640 x 480 with no assumption on the minimum detectable face size.
KAUST Department:
Visual Computing Center (VCC); Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division; Electrical Engineering Program
Publisher:
arXiv
Issue Date:
20-Jul-2017
ARXIV:
arXiv:1707.06330
Type:
Preprint
Additional Links:
http://arxiv.org/abs/1707.06330v1; http://arxiv.org/pdf/1707.06330v1
Appears in Collections:
Other/General Submission; Electrical Engineering Program; Visual Computing Center (VCC); Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorBai, Yanchengen
dc.contributor.authorGhanem, Bernarden
dc.date.accessioned2017-12-28T07:32:14Z-
dc.date.available2017-12-28T07:32:14Z-
dc.date.issued2017-07-20en
dc.identifier.urihttp://hdl.handle.net/10754/626520-
dc.description.abstractFace detection is a fundamental problem in computer vision. It is still a challenging task in unconstrained conditions due to significant variations in scale, pose, expressions, and occlusion. In this paper, we propose a multi-branch fully convolutional network (MB-FCN) for face detection, which considers both efficiency and effectiveness in the design process. Our MB-FCN detector can deal with faces at all scale ranges with only a single pass through the backbone network. As such, our MB-FCN model saves computation and thus is more efficient, compared to previous methods that make multiple passes. For each branch, the specific skip connections of the convolutional feature maps at different layers are exploited to represent faces in specific scale ranges. Specifically, small faces can be represented with both shallow fine-grained and deep powerful coarse features. With this representation, superior improvement in performance is registered for the task of detecting small faces. We test our MB-FCN detector on two public face detection benchmarks, including FDDB and WIDER FACE. Extensive experiments show that our detector outperforms state-of-the-art methods on all these datasets in general and by a substantial margin on the most challenging among them (e.g. WIDER FACE Hard subset). Also, MB-FCN runs at 15 FPS on a GPU for images of size 640 x 480 with no assumption on the minimum detectable face size.en
dc.publisherarXiven
dc.relation.urlhttp://arxiv.org/abs/1707.06330v1en
dc.relation.urlhttp://arxiv.org/pdf/1707.06330v1en
dc.rightsArchived with thanks to arXiven
dc.titleMulti-Branch Fully Convolutional Network for Face Detectionen
dc.typePreprinten
dc.contributor.departmentVisual Computing Center (VCC)en
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.contributor.departmentElectrical Engineering Programen
dc.eprint.versionPre-printen
dc.identifier.arxividarXiv:1707.06330en
kaust.authorBai, Yanchengen
kaust.authorGhanem, Bernarden
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.