Counting People Based on Linear, Weighted, and Local Random Forests

Helia Farhood; Xiangjian He; Wenjing Jia; Michael Blumenstein; Hanhui Li

doi:10.1109/DICTA.2017.8227421

Counting People Based on Linear, Weighted, and Local Random Forests

Helia Farhood
, Xiangjian He
, Wenjing Jia
, Michael Blumenstein
, Hanhui Li

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

10 Citations (Scopus)

Abstract

Recently, many works have been published for counting people. However, when being applied to real-world train station videos, they have exposed many limitations due to problems such as low resolution, heavy occlusion, various density levels and perspective distortions. In this paper, following the recent trend of regression-based density estimation, we present a linear regression approach based on local Random Forests for counting either standing or moving people on station platforms. By dividing each frame into sub-windows and extracting features with ground truth densities as well as learned weights, we perform a linear transformation for counting people to overcome the perspective problems of the existing patch-based approaches. We present improvements against several recent baselines on the UCSD dataset and a dataset of CCTV videos taken from a train station. We also show improvements in speed compared with the state-of-the-art models based on detection and Deep Learning.

Original language	English
Title of host publication	DICTA 2017 - 2017 International Conference on Digital Image Computing
Subtitle of host publication	Techniques and Applications
Editors	Yi Guo, Hongdong Li, Weidong Tom Cai, Manzur Murshed, Zhiyong Wang, Junbin Gao, David Dagan Feng
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1-7
Number of pages	7
ISBN (Electronic)	9781538628393
DOIs	https://doi.org/10.1109/DICTA.2017.8227421
Publication status	Published - 19 Dec 2017
Externally published	Yes
Event	19th International Conference on Digital Image Computing: Techniques and Applications, DICTA 2017 - Sydney, Australia Duration: 29 Nov 2017 → 1 Dec 2017

Publication series

Name	DICTA 2017 - 2017 International Conference on Digital Image Computing: Techniques and Applications
Volume	2017-December

Conference

Conference	19th International Conference on Digital Image Computing: Techniques and Applications, DICTA 2017
Country/Territory	Australia
City	Sydney
Period	29/11/17 → 1/12/17

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 15 Life on Land

Free Keywords

crowd counting
Density estimation
linear regression
Random Forest

ASJC Scopus subject areas

Computer Science Applications
Signal Processing

Access to Document

10.1109/DICTA.2017.8227421

Cite this

Farhood, H., He, X., Jia, W., Blumenstein, M., & Li, H. (2017). Counting People Based on Linear, Weighted, and Local Random Forests. In Y. Guo, H. Li, W. T. Cai, M. Murshed, Z. Wang, J. Gao, & D. D. Feng (Eds.), DICTA 2017 - 2017 International Conference on Digital Image Computing: Techniques and Applications (pp. 1-7). (DICTA 2017 - 2017 International Conference on Digital Image Computing: Techniques and Applications; Vol. 2017-December). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/DICTA.2017.8227421

Farhood, Helia ; He, Xiangjian ; Jia, Wenjing et al. / Counting People Based on Linear, Weighted, and Local Random Forests. DICTA 2017 - 2017 International Conference on Digital Image Computing: Techniques and Applications. editor / Yi Guo ; Hongdong Li ; Weidong Tom Cai ; Manzur Murshed ; Zhiyong Wang ; Junbin Gao ; David Dagan Feng. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 1-7 (DICTA 2017 - 2017 International Conference on Digital Image Computing: Techniques and Applications).

@inproceedings{737edcd8b6854346af6e643e7b190442,

title = "Counting People Based on Linear, Weighted, and Local Random Forests",

abstract = "Recently, many works have been published for counting people. However, when being applied to real-world train station videos, they have exposed many limitations due to problems such as low resolution, heavy occlusion, various density levels and perspective distortions. In this paper, following the recent trend of regression-based density estimation, we present a linear regression approach based on local Random Forests for counting either standing or moving people on station platforms. By dividing each frame into sub-windows and extracting features with ground truth densities as well as learned weights, we perform a linear transformation for counting people to overcome the perspective problems of the existing patch-based approaches. We present improvements against several recent baselines on the UCSD dataset and a dataset of CCTV videos taken from a train station. We also show improvements in speed compared with the state-of-the-art models based on detection and Deep Learning.",

keywords = "crowd counting, Density estimation, linear regression, Random Forest",

author = "Helia Farhood and Xiangjian He and Wenjing Jia and Michael Blumenstein and Hanhui Li",

note = "Publisher Copyright: {\textcopyright} 2017 IEEE.; 19th International Conference on Digital Image Computing: Techniques and Applications, DICTA 2017 ; Conference date: 29-11-2017 Through 01-12-2017",

year = "2017",

month = dec,

day = "19",

doi = "10.1109/DICTA.2017.8227421",

language = "English",

series = "DICTA 2017 - 2017 International Conference on Digital Image Computing: Techniques and Applications",

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

pages = "1--7",

editor = "Yi Guo and Hongdong Li and Cai, \{Weidong Tom\} and Manzur Murshed and Zhiyong Wang and Junbin Gao and Feng, \{David Dagan\}",

booktitle = "DICTA 2017 - 2017 International Conference on Digital Image Computing",

address = "United States",

}

Farhood, H, He, X, Jia, W, Blumenstein, M & Li, H 2017, Counting People Based on Linear, Weighted, and Local Random Forests. in Y Guo, H Li, WT Cai, M Murshed, Z Wang, J Gao & DD Feng (eds), DICTA 2017 - 2017 International Conference on Digital Image Computing: Techniques and Applications. DICTA 2017 - 2017 International Conference on Digital Image Computing: Techniques and Applications, vol. 2017-December, Institute of Electrical and Electronics Engineers Inc., pp. 1-7, 19th International Conference on Digital Image Computing: Techniques and Applications, DICTA 2017, Sydney, Australia, 29/11/17. https://doi.org/10.1109/DICTA.2017.8227421

Counting People Based on Linear, Weighted, and Local Random Forests. / Farhood, Helia; He, Xiangjian; Jia, Wenjing et al.
DICTA 2017 - 2017 International Conference on Digital Image Computing: Techniques and Applications. ed. / Yi Guo; Hongdong Li; Weidong Tom Cai; Manzur Murshed; Zhiyong Wang; Junbin Gao; David Dagan Feng. Institute of Electrical and Electronics Engineers Inc., 2017. p. 1-7 (DICTA 2017 - 2017 International Conference on Digital Image Computing: Techniques and Applications; Vol. 2017-December).

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Counting People Based on Linear, Weighted, and Local Random Forests

AU - Farhood, Helia

AU - He, Xiangjian

AU - Jia, Wenjing

AU - Blumenstein, Michael

AU - Li, Hanhui

PY - 2017/12/19

Y1 - 2017/12/19

N2 - Recently, many works have been published for counting people. However, when being applied to real-world train station videos, they have exposed many limitations due to problems such as low resolution, heavy occlusion, various density levels and perspective distortions. In this paper, following the recent trend of regression-based density estimation, we present a linear regression approach based on local Random Forests for counting either standing or moving people on station platforms. By dividing each frame into sub-windows and extracting features with ground truth densities as well as learned weights, we perform a linear transformation for counting people to overcome the perspective problems of the existing patch-based approaches. We present improvements against several recent baselines on the UCSD dataset and a dataset of CCTV videos taken from a train station. We also show improvements in speed compared with the state-of-the-art models based on detection and Deep Learning.

AB - Recently, many works have been published for counting people. However, when being applied to real-world train station videos, they have exposed many limitations due to problems such as low resolution, heavy occlusion, various density levels and perspective distortions. In this paper, following the recent trend of regression-based density estimation, we present a linear regression approach based on local Random Forests for counting either standing or moving people on station platforms. By dividing each frame into sub-windows and extracting features with ground truth densities as well as learned weights, we perform a linear transformation for counting people to overcome the perspective problems of the existing patch-based approaches. We present improvements against several recent baselines on the UCSD dataset and a dataset of CCTV videos taken from a train station. We also show improvements in speed compared with the state-of-the-art models based on detection and Deep Learning.

KW - crowd counting

KW - Density estimation

KW - linear regression

KW - Random Forest

UR - https://www.scopus.com/pages/publications/85048287608

U2 - 10.1109/DICTA.2017.8227421

DO - 10.1109/DICTA.2017.8227421

M3 - Conference contribution

AN - SCOPUS:85048287608

T3 - DICTA 2017 - 2017 International Conference on Digital Image Computing: Techniques and Applications

SP - 1

EP - 7

BT - DICTA 2017 - 2017 International Conference on Digital Image Computing

A2 - Guo, Yi

A2 - Li, Hongdong

A2 - Cai, Weidong Tom

A2 - Murshed, Manzur

A2 - Wang, Zhiyong

A2 - Gao, Junbin

A2 - Feng, David Dagan

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 19th International Conference on Digital Image Computing: Techniques and Applications, DICTA 2017

Y2 - 29 November 2017 through 1 December 2017

ER -

Farhood H, He X, Jia W, Blumenstein M, Li H. Counting People Based on Linear, Weighted, and Local Random Forests. In Guo Y, Li H, Cai WT, Murshed M, Wang Z, Gao J, Feng DD, editors, DICTA 2017 - 2017 International Conference on Digital Image Computing: Techniques and Applications. Institute of Electrical and Electronics Engineers Inc. 2017. p. 1-7. (DICTA 2017 - 2017 International Conference on Digital Image Computing: Techniques and Applications). doi: 10.1109/DICTA.2017.8227421

Counting People Based on Linear, Weighted, and Local Random Forests

Abstract

Publication series

Conference

UN SDGs

Free Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this