Granular-based dense crowd density estimation

Kok Ven Jyn , Chee Seng Chan

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Dense crowd density estimation is one of the fundamental tasks in crowd analysis. While tremendous progress has been made to understand crowd scenes along with the rise of Convolutional Neural Networks (CNNs), research work on dense crowd density estimation is still an ongoing process. In this paper, we propose a novel approach to learn discriminative crowd features from granules, that conforms to the outline between crowd and background (i.e. non-crowd) regions, for density estimation. It shows that by studying the inner statistics of granules for density estimation, this approach is adaptive to arbitrary distribution of crowd (i.e. scene independent). Multiple features fusion is proposed to learn discriminative crowd features from granules. This is to be used as description of the crowd where a direct mapping between the features and crowd density is learned. Extensive experiments on public benchmark datasets demonstrate the effectiveness of our novel approach for scene independent dense crowd density estimation.

Original languageEnglish
Pages (from-to)1-20
Number of pages20
JournalMultimedia Tools and Applications
DOIs
Publication statusAccepted/In press - 5 Dec 2017

Fingerprint

Fusion reactions
Statistics
Neural networks
Experiments

Keywords

  • Dense crowd analysis
  • Density estimation
  • Texture features
  • Visual surveillance

ASJC Scopus subject areas

  • Software
  • Media Technology
  • Hardware and Architecture
  • Computer Networks and Communications

Cite this

Granular-based dense crowd density estimation. / Ven Jyn , Kok; Chan, Chee Seng.

In: Multimedia Tools and Applications, 05.12.2017, p. 1-20.

Research output: Contribution to journalArticle

@article{6f1f0be977cd49d1878a6950b4a2c8da,
title = "Granular-based dense crowd density estimation",
abstract = "Dense crowd density estimation is one of the fundamental tasks in crowd analysis. While tremendous progress has been made to understand crowd scenes along with the rise of Convolutional Neural Networks (CNNs), research work on dense crowd density estimation is still an ongoing process. In this paper, we propose a novel approach to learn discriminative crowd features from granules, that conforms to the outline between crowd and background (i.e. non-crowd) regions, for density estimation. It shows that by studying the inner statistics of granules for density estimation, this approach is adaptive to arbitrary distribution of crowd (i.e. scene independent). Multiple features fusion is proposed to learn discriminative crowd features from granules. This is to be used as description of the crowd where a direct mapping between the features and crowd density is learned. Extensive experiments on public benchmark datasets demonstrate the effectiveness of our novel approach for scene independent dense crowd density estimation.",
keywords = "Dense crowd analysis, Density estimation, Texture features, Visual surveillance",
author = "{Ven Jyn }, Kok and Chan, {Chee Seng}",
year = "2017",
month = "12",
day = "5",
doi = "10.1007/s11042-017-5418-y",
language = "English",
pages = "1--20",
journal = "Multimedia Tools and Applications",
issn = "1380-7501",
publisher = "Springer Netherlands",

}

TY - JOUR

T1 - Granular-based dense crowd density estimation

AU - Ven Jyn , Kok

AU - Chan, Chee Seng

PY - 2017/12/5

Y1 - 2017/12/5

N2 - Dense crowd density estimation is one of the fundamental tasks in crowd analysis. While tremendous progress has been made to understand crowd scenes along with the rise of Convolutional Neural Networks (CNNs), research work on dense crowd density estimation is still an ongoing process. In this paper, we propose a novel approach to learn discriminative crowd features from granules, that conforms to the outline between crowd and background (i.e. non-crowd) regions, for density estimation. It shows that by studying the inner statistics of granules for density estimation, this approach is adaptive to arbitrary distribution of crowd (i.e. scene independent). Multiple features fusion is proposed to learn discriminative crowd features from granules. This is to be used as description of the crowd where a direct mapping between the features and crowd density is learned. Extensive experiments on public benchmark datasets demonstrate the effectiveness of our novel approach for scene independent dense crowd density estimation.

AB - Dense crowd density estimation is one of the fundamental tasks in crowd analysis. While tremendous progress has been made to understand crowd scenes along with the rise of Convolutional Neural Networks (CNNs), research work on dense crowd density estimation is still an ongoing process. In this paper, we propose a novel approach to learn discriminative crowd features from granules, that conforms to the outline between crowd and background (i.e. non-crowd) regions, for density estimation. It shows that by studying the inner statistics of granules for density estimation, this approach is adaptive to arbitrary distribution of crowd (i.e. scene independent). Multiple features fusion is proposed to learn discriminative crowd features from granules. This is to be used as description of the crowd where a direct mapping between the features and crowd density is learned. Extensive experiments on public benchmark datasets demonstrate the effectiveness of our novel approach for scene independent dense crowd density estimation.

KW - Dense crowd analysis

KW - Density estimation

KW - Texture features

KW - Visual surveillance

UR - http://www.scopus.com/inward/record.url?scp=85037613091&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85037613091&partnerID=8YFLogxK

U2 - 10.1007/s11042-017-5418-y

DO - 10.1007/s11042-017-5418-y

M3 - Article

AN - SCOPUS:85037613091

SP - 1

EP - 20

JO - Multimedia Tools and Applications

JF - Multimedia Tools and Applications

SN - 1380-7501

ER -