Workshop on

Crowd Analysis and Applications: Simulations Meet Video Analytics

Organizing Committee: Robot Vision Team, Kingston University London, United Kingdom


Studying human behaviour in crowds is of great scientific interest in a number of research disciplines, including computer simulations, graphics, and video analytics. Crowd analysis lends itself to a number of applications including public safety, ergonomics and video surveillance, congestion and flow analysis, stampede avoidance, marketing and tourism. Research in crowd analysis is at least two decades old and it has grown exponentially, supported by over 500 scientific publications (Scopus statistics). The precise analysis of complex behaviour in crowds must be based on real data, hard to obtain, and unrealistic to label. Computer simulations are therefore needed to create realistic examples of crowd dynamics that can then be employed to train machine learning algorithms to classify and recognise behaviour in very busy scenes, potentially populated by large numbers of people.

The main focus of this workshop is on crowd analysis and its applications using simulations, computer vision and machine learning techniques, in particular following approaches based on deep learning. We welcome original submissions on:

  • Counting, Localization and Density Estimation of real scenes based on machine learnt models
  • Activity Recognition and Behaviour Analysis based on models of interactions
  • Simulation and Modelling of very crowded scenes in complex environments
  • Abnormal Event Detection based on the modelling of abrupt motion models
  • Management, Security and Surveillance of large crowds
  • Person Flow Analysis based on dynamic models
  • Improving on Person Tracking using deep learning methods
  • Use of Generative methods for the creation of realistic crowd simulations

The above topics are indicative, any original work related to crowd analysis is also encouraged.

Guides for Authors:

This workshop seeks original, high-quality papers in all areas related to crowd analysis and related applications. Full paper submissions are limited to 4 to 6 pages in length, including references and must be submitted electronically using the provided LaTeX conference template or Microsoft Word conference template. Authors should submit their manuscripts in English, addressing one or several of the above-mentioned conference areas. To facilitate the double-blind paper evaluation method, authors are kindly requested to prepare their paper without any reference to any of the author’s details and acknowledgements. One of the authors will have to register and present their work at the workshop. Each presenter will have 15 minutes for their presentation, followed by 5 minutes reserved to a Q&A session. Papers should be submitted in the PDF file format through EasyChair provided below:

Important Dates:

Submission deadline: 30 May 2019

Notification for Acceptance: 10 June 2019

Workshop Date: 3 July 2019


Program Committee:

Dr. Ikushi Yoda

National Institute of Advanced Industrial Science and Technology (AIST), Japan

Dr. Nuria Pelechano

Universitat Politecnica de Catalunya (UPC), Barcelona, Spain

Prof. Keith Still

Manchester Metropolitan University, United Kingdom

Prof. Soraia Raupp Musse

Computing Science Department, Pontifical Catholic University, Brazil

Prof. Shaogang Gong

School of Electronic Eng & Comp Science, Queen Mary University, United Kingdom

Dr. Lucio Marcenaro

DITEN, University of Genova

Genoa, Italy

Dr. Chee Seng Chan

Dept of Artificial Intelligence, University of Malaya, Kuala Lumpur, Malaysia

Dr. Rob Dupre

VCA Technology Inc, London, United Kingdom

Dr. Thomas Lagkas

Dept of Computer Science, The University of Sheffield, United Kingdom

Dr. Panagiotis Sarigiannidis

Dept of Informatics and Telecommunications Eng, University of Western Macedonia, Greece

Dr. Lim Mei Kuan

School of IT, Monash University, Kuala Lumpur, Malaysia

Dr. Oksana Koltsova

Legion Pedestrian Simulation, Bentley Systems, United Kingdom

Dr. Francois Bremond

Institut National de Informatique et en Automatique (INRIA) , Rocquencourt, France