Score-specific Non-maximum Suppression and Coexistence Prior for Multi-scale Face Detection

Tianpeng Wu; Dong Liang; Jiaxing Pan; Han Sun; Bin Kang; Shunrichi Kaneko; Huiyu Zhou

doi:10.1109/ICASSP.2019.8682567

Score-specific Non-maximum Suppression and Coexistence Prior for Multi-scale Face Detection

Tianpeng Wu, Dong Liang, Jiaxing Pan, Han Sun, Bin Kang, Shunrichi Kaneko, Huiyu Zhou

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

2 Citations (Scopus)

Abstract

Face detection is an ultimate component to support various visual facial related tasks. However, detecting faces with extremely low resolution or high occlusion is still an open problem. In this paper, we propose a two-step general approach to refine the performance of modern face detectors according to human's high-level context-aware ability. First, we propose Score-specific Non-Maximum Suppression (SNMS) to preserve overlapped faces. Second, we consider the coexistence prior among faces in the scene, which could raise the sensitivity of face detection in the crowd. When integrating our approach to the existing face detectors, most of them have better results on a challenging benchmark (WIDER FACE) and a newly proposed dataset (Faces in Crowd, FIC) made by us. Codes are available on https://github.com/AIoTP/SNMSandCoexistence.

Original language	English
Title of host publication	2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1957-1961
Number of pages	5
ISBN (Electronic)	9781479981311
DOIs	https://doi.org/10.1109/ICASSP.2019.8682567
Publication status	Published - 17 Apr 2019
Externally published	Yes
Event	IEEE International Conference on Acoustics, Speech, and Signal Processing 2019 - Brighton, United Kingdom Duration: 12 May 2019 → 17 May 2019 https://doi.org/10.1109/ICASSP35589.2019 https://2019.ieeeicassp.org/

Publication series

Name	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume	2019-May
ISSN (Print)	1520-6149

Conference

Conference	IEEE International Conference on Acoustics, Speech, and Signal Processing 2019
Abbreviated title	ICASSP 2019
Country/Territory	United Kingdom
City	Brighton
Period	12/05/2019 → 17/05/2019
Internet address	https://doi.org/10.1109/ICASSP35589.2019 https://2019.ieeeicassp.org/

Bibliographical note

Funding Information:
Thiswork issupported bythe NationalKeyR&D Program of China under Grant 2017YFB0802300, the National Natural Science Foundation of China (Grants No 61601223， 61801242), the Natural Science Foundation of Jiangsu Province (Grants No BK20150756)，China Postdoctoral Science Foundation (Grants No 2015M580427)，and the Fundamental Research Funds for the Central Universities ( No. NS2016091).

Publisher Copyright:
© 2019 IEEE.

Copyright:
Copyright 2019 Elsevier B.V., All rights reserved.

Keywords

Coexistence prior
Contextual information
Face detection
Score-specific NMS

ASJC Scopus subject areas

Software
Signal Processing
Electrical and Electronic Engineering

Access to Document

10.1109/ICASSP.2019.8682567Licence: Unspecified

Cite this

Wu, T., Liang, D., Pan, J., Sun, H., Kang, B., Kaneko, S., & Zhou, H. (2019). Score-specific Non-maximum Suppression and Coexistence Prior for Multi-scale Face Detection. In 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings (pp. 1957-1961). Article 8682567 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2019-May). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICASSP.2019.8682567

Wu, Tianpeng ; Liang, Dong ; Pan, Jiaxing et al. / Score-specific Non-maximum Suppression and Coexistence Prior for Multi-scale Face Detection. 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 1957-1961 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

@inproceedings{3db5342dbc2d487c9f6bbf76ef1ecc3c,

title = "Score-specific Non-maximum Suppression and Coexistence Prior for Multi-scale Face Detection",

abstract = "Face detection is an ultimate component to support various visual facial related tasks. However, detecting faces with extremely low resolution or high occlusion is still an open problem. In this paper, we propose a two-step general approach to refine the performance of modern face detectors according to human's high-level context-aware ability. First, we propose Score-specific Non-Maximum Suppression (SNMS) to preserve overlapped faces. Second, we consider the coexistence prior among faces in the scene, which could raise the sensitivity of face detection in the crowd. When integrating our approach to the existing face detectors, most of them have better results on a challenging benchmark (WIDER FACE) and a newly proposed dataset (Faces in Crowd, FIC) made by us. Codes are available on https://github.com/AIoTP/SNMSandCoexistence.",

keywords = "Coexistence prior, Contextual information, Face detection, Score-specific NMS",

author = "Tianpeng Wu and Dong Liang and Jiaxing Pan and Han Sun and Bin Kang and Shunrichi Kaneko and Huiyu Zhou",

note = "Funding Information: Thiswork issupported bythe NationalKeyR&D Program of China under Grant 2017YFB0802300, the National Natural Science Foundation of China (Grants No 61601223， 61801242), the Natural Science Foundation of Jiangsu Province (Grants No BK20150756)，China Postdoctoral Science Foundation (Grants No 2015M580427)，and the Fundamental Research Funds for the Central Universities ( No. NS2016091). Publisher Copyright: {\textcopyright} 2019 IEEE. Copyright: Copyright 2019 Elsevier B.V., All rights reserved.; IEEE International Conference on Acoustics, Speech, and Signal Processing 2019, ICASSP 2019 ; Conference date: 12-05-2019 Through 17-05-2019",

year = "2019",

month = apr,

day = "17",

doi = "10.1109/ICASSP.2019.8682567",

language = "English",

series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1957--1961",

booktitle = "2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings",

address = "United States",

url = "https://doi.org/10.1109/ICASSP35589.2019, https://2019.ieeeicassp.org/",

}

Wu, T, Liang, D, Pan, J, Sun, H, Kang, B, Kaneko, S & Zhou, H 2019, Score-specific Non-maximum Suppression and Coexistence Prior for Multi-scale Face Detection. in 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings., 8682567, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, vol. 2019-May, Institute of Electrical and Electronics Engineers Inc., pp. 1957-1961, IEEE International Conference on Acoustics, Speech, and Signal Processing 2019, Brighton, United Kingdom, 12/05/2019. https://doi.org/10.1109/ICASSP.2019.8682567

Score-specific Non-maximum Suppression and Coexistence Prior for Multi-scale Face Detection. / Wu, Tianpeng; Liang, Dong; Pan, Jiaxing et al.
2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. p. 1957-1961 8682567 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2019-May).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Score-specific Non-maximum Suppression and Coexistence Prior for Multi-scale Face Detection

AU - Wu, Tianpeng

AU - Liang, Dong

AU - Pan, Jiaxing

AU - Sun, Han

AU - Kang, Bin

AU - Kaneko, Shunrichi

AU - Zhou, Huiyu

N1 - Funding Information: Thiswork issupported bythe NationalKeyR&D Program of China under Grant 2017YFB0802300, the National Natural Science Foundation of China (Grants No 61601223， 61801242), the Natural Science Foundation of Jiangsu Province (Grants No BK20150756)，China Postdoctoral Science Foundation (Grants No 2015M580427)，and the Fundamental Research Funds for the Central Universities ( No. NS2016091). Publisher Copyright: © 2019 IEEE. Copyright: Copyright 2019 Elsevier B.V., All rights reserved.

PY - 2019/4/17

Y1 - 2019/4/17

N2 - Face detection is an ultimate component to support various visual facial related tasks. However, detecting faces with extremely low resolution or high occlusion is still an open problem. In this paper, we propose a two-step general approach to refine the performance of modern face detectors according to human's high-level context-aware ability. First, we propose Score-specific Non-Maximum Suppression (SNMS) to preserve overlapped faces. Second, we consider the coexistence prior among faces in the scene, which could raise the sensitivity of face detection in the crowd. When integrating our approach to the existing face detectors, most of them have better results on a challenging benchmark (WIDER FACE) and a newly proposed dataset (Faces in Crowd, FIC) made by us. Codes are available on https://github.com/AIoTP/SNMSandCoexistence.

AB - Face detection is an ultimate component to support various visual facial related tasks. However, detecting faces with extremely low resolution or high occlusion is still an open problem. In this paper, we propose a two-step general approach to refine the performance of modern face detectors according to human's high-level context-aware ability. First, we propose Score-specific Non-Maximum Suppression (SNMS) to preserve overlapped faces. Second, we consider the coexistence prior among faces in the scene, which could raise the sensitivity of face detection in the crowd. When integrating our approach to the existing face detectors, most of them have better results on a challenging benchmark (WIDER FACE) and a newly proposed dataset (Faces in Crowd, FIC) made by us. Codes are available on https://github.com/AIoTP/SNMSandCoexistence.

KW - Coexistence prior

KW - Contextual information

KW - Face detection

KW - Score-specific NMS

U2 - 10.1109/ICASSP.2019.8682567

DO - 10.1109/ICASSP.2019.8682567

M3 - Conference contribution

AN - SCOPUS:85068965039

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 1957

EP - 1961

BT - 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - IEEE International Conference on Acoustics, Speech, and Signal Processing 2019

Y2 - 12 May 2019 through 17 May 2019

ER -

Wu T, Liang D, Pan J, Sun H, Kang B, Kaneko S et al. Score-specific Non-maximum Suppression and Coexistence Prior for Multi-scale Face Detection. In 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2019. p. 1957-1961. 8682567. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). doi: 10.1109/ICASSP.2019.8682567

Score-specific Non-maximum Suppression and Coexistence Prior for Multi-scale Face Detection

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this