A Deep Neural Network Approach for Top View People Detection and Counting

Misbah Ahmad; Imran Ahmed; Kaleem Ullah; Maaz Ahmad

doi:10.1109/UEMCON47517.2019.8993109

A Deep Neural Network Approach for Top View People Detection and Counting

Misbah Ahmad, Imran Ahmed, Kaleem Ullah, Maaz Ahmad

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

36 Citations (Scopus)

Abstract

People detection and counting is considered as one of the important application in video surveillance. Various computer vision and deep learning based methods have been developed which aim to provide efficient and accurate people detection/counting results using frontal view data sets. Furthermore, there are many challenges which occurs while detecting people including occlusion, perspective distortion, variations in human body pose, size and orientation, these challenges effect the results of developed detection and counting models. In this work, a deep neural network approach i.e. SSD (Single Shot multi-box Detector) is explored for people detection and counting. SSD model is used for detection and counting of people from significantly different viewpoint i.e. top view. To the extent of our knowledge, this is the first attempt to use deep neural network based model for top view people detection and counting. Furthermore, the impact of frontal view trained SSD model on top view test images is also discussed. The experimental results show the effectiveness of deep learning model by achieving promising results with average TPR of 95% and TPR 94.42% for indoor and outdoor environments respectively.

Original language	English (UK)
Title of host publication	2019 IEEE 10th Annual Ubiquitous Computing, Electronics and Mobile Communication Conference, UEMCON 2019
Editors	Satyajit Chakrabarti, Himadri Nath Saha
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1082-1088
Number of pages	7
ISBN (Electronic)	9781728138855
DOIs	https://doi.org/10.1109/UEMCON47517.2019.8993109
Publication status	Published - Oct 2019
Externally published	Yes
Event	10th IEEE Annual Ubiquitous Computing, Electronics and Mobile Communication Conference, UEMCON 2019 - New York City, United States Duration: 10 Oct 2019 → 12 Oct 2019

Publication series

Name	2019 IEEE 10th Annual Ubiquitous Computing, Electronics and Mobile Communication Conference, UEMCON 2019

Conference

Conference	10th IEEE Annual Ubiquitous Computing, Electronics and Mobile Communication Conference, UEMCON 2019
Country/Territory	United States
City	New York City
Period	10/10/19 → 12/10/19

Keywords

Deep Neural Network
Single Shot Detector
Top view detection
people counting

Access to Document

10.1109/UEMCON47517.2019.8993109

Cite this

Ahmad, M., Ahmed, I., Ullah, K., & Ahmad, M. (2019). A Deep Neural Network Approach for Top View People Detection and Counting. In S. Chakrabarti, & H. N. Saha (Eds.), 2019 IEEE 10th Annual Ubiquitous Computing, Electronics and Mobile Communication Conference, UEMCON 2019 (pp. 1082-1088). Article 8993109 (2019 IEEE 10th Annual Ubiquitous Computing, Electronics and Mobile Communication Conference, UEMCON 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/UEMCON47517.2019.8993109

Ahmad, Misbah ; Ahmed, Imran ; Ullah, Kaleem et al. / A Deep Neural Network Approach for Top View People Detection and Counting. 2019 IEEE 10th Annual Ubiquitous Computing, Electronics and Mobile Communication Conference, UEMCON 2019. editor / Satyajit Chakrabarti ; Himadri Nath Saha. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 1082-1088 (2019 IEEE 10th Annual Ubiquitous Computing, Electronics and Mobile Communication Conference, UEMCON 2019).

@inproceedings{b52e70c27a8e46bbb22403734ade1baa,

title = "A Deep Neural Network Approach for Top View People Detection and Counting",

abstract = "People detection and counting is considered as one of the important application in video surveillance. Various computer vision and deep learning based methods have been developed which aim to provide efficient and accurate people detection/counting results using frontal view data sets. Furthermore, there are many challenges which occurs while detecting people including occlusion, perspective distortion, variations in human body pose, size and orientation, these challenges effect the results of developed detection and counting models. In this work, a deep neural network approach i.e. SSD (Single Shot multi-box Detector) is explored for people detection and counting. SSD model is used for detection and counting of people from significantly different viewpoint i.e. top view. To the extent of our knowledge, this is the first attempt to use deep neural network based model for top view people detection and counting. Furthermore, the impact of frontal view trained SSD model on top view test images is also discussed. The experimental results show the effectiveness of deep learning model by achieving promising results with average TPR of 95\% and TPR 94.42\% for indoor and outdoor environments respectively.",

keywords = "Deep Neural Network, Single Shot Detector, Top view detection, people counting",

author = "Misbah Ahmad and Imran Ahmed and Kaleem Ullah and Maaz Ahmad",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; 10th IEEE Annual Ubiquitous Computing, Electronics and Mobile Communication Conference, UEMCON 2019 ; Conference date: 10-10-2019 Through 12-10-2019",

year = "2019",

month = oct,

doi = "10.1109/UEMCON47517.2019.8993109",

language = "English (UK)",

series = "2019 IEEE 10th Annual Ubiquitous Computing, Electronics and Mobile Communication Conference, UEMCON 2019",

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

pages = "1082--1088",

editor = "Satyajit Chakrabarti and Saha, \{Himadri Nath\}",

booktitle = "2019 IEEE 10th Annual Ubiquitous Computing, Electronics and Mobile Communication Conference, UEMCON 2019",

address = "United States",

}

Ahmad, M, Ahmed, I, Ullah, K & Ahmad, M 2019, A Deep Neural Network Approach for Top View People Detection and Counting. in S Chakrabarti & HN Saha (eds), 2019 IEEE 10th Annual Ubiquitous Computing, Electronics and Mobile Communication Conference, UEMCON 2019., 8993109, 2019 IEEE 10th Annual Ubiquitous Computing, Electronics and Mobile Communication Conference, UEMCON 2019, Institute of Electrical and Electronics Engineers Inc., pp. 1082-1088, 10th IEEE Annual Ubiquitous Computing, Electronics and Mobile Communication Conference, UEMCON 2019, New York City, United States, 10/10/19. https://doi.org/10.1109/UEMCON47517.2019.8993109

A Deep Neural Network Approach for Top View People Detection and Counting. / Ahmad, Misbah; Ahmed, Imran; Ullah, Kaleem et al.
2019 IEEE 10th Annual Ubiquitous Computing, Electronics and Mobile Communication Conference, UEMCON 2019. ed. / Satyajit Chakrabarti; Himadri Nath Saha. Institute of Electrical and Electronics Engineers Inc., 2019. p. 1082-1088 8993109 (2019 IEEE 10th Annual Ubiquitous Computing, Electronics and Mobile Communication Conference, UEMCON 2019).

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

TY - GEN

T1 - A Deep Neural Network Approach for Top View People Detection and Counting

AU - Ahmad, Misbah

AU - Ahmed, Imran

AU - Ullah, Kaleem

AU - Ahmad, Maaz

PY - 2019/10

Y1 - 2019/10

N2 - People detection and counting is considered as one of the important application in video surveillance. Various computer vision and deep learning based methods have been developed which aim to provide efficient and accurate people detection/counting results using frontal view data sets. Furthermore, there are many challenges which occurs while detecting people including occlusion, perspective distortion, variations in human body pose, size and orientation, these challenges effect the results of developed detection and counting models. In this work, a deep neural network approach i.e. SSD (Single Shot multi-box Detector) is explored for people detection and counting. SSD model is used for detection and counting of people from significantly different viewpoint i.e. top view. To the extent of our knowledge, this is the first attempt to use deep neural network based model for top view people detection and counting. Furthermore, the impact of frontal view trained SSD model on top view test images is also discussed. The experimental results show the effectiveness of deep learning model by achieving promising results with average TPR of 95% and TPR 94.42% for indoor and outdoor environments respectively.

AB - People detection and counting is considered as one of the important application in video surveillance. Various computer vision and deep learning based methods have been developed which aim to provide efficient and accurate people detection/counting results using frontal view data sets. Furthermore, there are many challenges which occurs while detecting people including occlusion, perspective distortion, variations in human body pose, size and orientation, these challenges effect the results of developed detection and counting models. In this work, a deep neural network approach i.e. SSD (Single Shot multi-box Detector) is explored for people detection and counting. SSD model is used for detection and counting of people from significantly different viewpoint i.e. top view. To the extent of our knowledge, this is the first attempt to use deep neural network based model for top view people detection and counting. Furthermore, the impact of frontal view trained SSD model on top view test images is also discussed. The experimental results show the effectiveness of deep learning model by achieving promising results with average TPR of 95% and TPR 94.42% for indoor and outdoor environments respectively.

KW - Deep Neural Network

KW - Single Shot Detector

KW - Top view detection

KW - people counting

UR - https://www.scopus.com/pages/publications/85080111813

U2 - 10.1109/UEMCON47517.2019.8993109

DO - 10.1109/UEMCON47517.2019.8993109

M3 - Conference contribution

AN - SCOPUS:85080111813

T3 - 2019 IEEE 10th Annual Ubiquitous Computing, Electronics and Mobile Communication Conference, UEMCON 2019

SP - 1082

EP - 1088

BT - 2019 IEEE 10th Annual Ubiquitous Computing, Electronics and Mobile Communication Conference, UEMCON 2019

A2 - Chakrabarti, Satyajit

A2 - Saha, Himadri Nath

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 10th IEEE Annual Ubiquitous Computing, Electronics and Mobile Communication Conference, UEMCON 2019

Y2 - 10 October 2019 through 12 October 2019

ER -

Ahmad M, Ahmed I, Ullah K, Ahmad M. A Deep Neural Network Approach for Top View People Detection and Counting. In Chakrabarti S, Saha HN, editors, 2019 IEEE 10th Annual Ubiquitous Computing, Electronics and Mobile Communication Conference, UEMCON 2019. Institute of Electrical and Electronics Engineers Inc. 2019. p. 1082-1088. 8993109. (2019 IEEE 10th Annual Ubiquitous Computing, Electronics and Mobile Communication Conference, UEMCON 2019). doi: 10.1109/UEMCON47517.2019.8993109

A Deep Neural Network Approach for Top View People Detection and Counting

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this