Motion estimation of high density crowd using fluid dynamics

ABSTRACT

Motion estimation (ME) being a fundamental process of crowd behavior analysis experienced real challenges at high densities due to visual ambiguities and occlusion problems etc. Various surveys reported in the past years summarize conventional ME methods for crowd behaviors at low/medium densities. In this paper, we focus on state-of-the-art fluid dynamics (FD) ME methods developed over the last one and the half-decade for high-density crowd analysis. A detailed discussion is provided on the development of FD ME methods explaining the strengths and weaknesses and viability of FD ME methods for anomaly detection at high crowd densities. Comprehensive experiments are performed comparing the performance of conventional and FD ME at varying crowd densities. Experimentation results show that conventional ME methods fail at high-density crowd whereas FD ME methods could estimate motion only at the global level. Still, research is required to meet the challenges of local ME at high crowd densities.

KEYWORDS:

• Motion estimation
• high-density crowd
• fluid dynamics
• particle advection
• FTLE
• flow visualization
• computational complexity
• local and global motion

Acknowledgements

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes on contributors

Muhammad Umer Farooq received B.S in Electrical Engineering from COMSATS Institute of Information Technology (2004) and Master from Universiti Teknologi PETRONAS (UTP) in 2013. He has more than ten years of industry job experience worked in areas of Mixed Signal circuit design and development, digital design using FPGA (Xilinx/Altera), and firmware development of various DSP processors, etc. Currently, he is working at Intel and pursuing his Ph.D. at the Center for Intelligent Signal and Imaging Research (CISIR) UTP. His research interests include behavior detection at the high-density crowd and deep learning.

Mohamed Naufal B. M. Saad received the master's degree from the Ecole Nationale Supérieure d'Ingénieurs de Limoges, France, and the Ph.D. degree in telecommunication from the Université de Limoges, France, in 2005. He is currently an Associate Professor with the Electrical and Electronic Engineering Department, Universiti Teknologi Petronas (UTP), Malaysia. He is a Core Research Member with the Center for Intelligent Signal and Imaging Research, UTP. His research interests include Neuro signal processing, medical imaging, and communication.

Associate Prof, Dr. Aamir Saeed Malik received B.S in Electrical Engineering from the University of Engineering & Technology, Lahore, Pakistan, M.S in Information & Communication and Ph.D. in Information & Mechatronics from Gwangju Institute of Science & Technology, South Korea. He has more than 15 years of research experience and is currently working as Associate Professor and Director of Biomedical Technology group at Universiti Teknologi PETRONAS. His research interests include biomedical signal & image processing, visual surveillance, remote sensing, and brain sciences.

Yasir Salih Ali (M'07) has received his Ph.D. in Electrical and Engineering from Universiti Teknologi PETRONAS, Malaysia, in 2015. He also received his BSc and MSc from the same university, in 2010 and 2011 respectively. He has worked as a teaching assistant at Universiti Teknologi Petronas in Malaysia from 2010 to 2012 teaching image processing implementation on MATLAB. He also worked as a visiting researcher in the Faculty of Engineering at Qatar University in 2013. Currently, he is an assistant professor in the Science and Technology Unit at Umm Al-Qura University, Makkah Saudi Arabia. His research interest includes computer vision, medical image, robotic vision, and signal processing.

Sultan Daud Khan is currently an Associate Professor in the Department of Computer Science, National University of Technology, Pakistan. He has published several papers in conferences and journals such as AVSS, IVCNZ, ICGIP, Neurocomputing, Journal of Cellular Automata, and IEEE Access. His research interests include crowd analysis, action recognition and localization, object detection, visual tracking, multi-camera, and airborne surveillance using deep learning techniques. He received the BSc (Hons) degree in Computer Engineering from the University of Engineering & Technology in 2005, MSc (Hons) in Electronics & Communication Engineering from Hanyang University, South Korea in 2010, and the Ph.D. degree in Computer Science from the University of Milano-Bicocca 2016. He is an active reviewer of prestigious journals, Neurocomputing, Pattern Recognition, IEEE IET signal processing, ACM Multimedia, IEEE Access, and ACM TOMM. He received the best reviewer award from Pattern Recognition in 2017.