University of Leicester
Browse

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

Download (815.78 kB)
conference contribution
posted on 2019-09-04, 15:07 authored by T Wu, D Liang, J Pan, H Sun, B Kang, S Kaneko, H Zhou
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.

Funding

This work is supported by the National Key R&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).

History

Citation

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, 2019, 2019-May, pp. 1957-1961

Author affiliation

/Organisation/COLLEGE OF SCIENCE AND ENGINEERING/Department of Informatics

Source

ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Brighton, United Kingdom

Version

  • AM (Accepted Manuscript)

Published in

ICASSP

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

issn

1520-6149

isbn

9781479981311

Copyright date

2019

Available date

2019-09-04

Publisher version

https://ieeexplore.ieee.org/document/8682567

Temporal coverage: start date

2019-05-12

Temporal coverage: end date

2019-05-17

Language

en

Usage metrics

    University of Leicester Publications

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC