Knowledge-based Approach to Crowd Analysis in Video Surveillance - KACAVIS


HRZZ - IP-01-2018

 
 

Knowledge-based Approach to Crowd Analysis in Video Surveillance - KACAVIS

Project summary

Recently, one of the most active research areas in computer vision is crowd scene analysis. Crowd modelling, crowd scene analysis and behavioural recognition are some of most challenging topics in computer vision and more generally in artificial intelligence. The scientific goal of the proposed research is the development of an experimental intelligent, knowledge-based system for crowd analysis and behaviour recognition based on visual information obtained by video surveillance systems. A proposal is made for a novel knowledge-based approach to modelling crowd scenes, which represents a fusion of agent-based and entity-based approaches, together with human common-sense modelling. The idea is to convert human (or expert) knowledge about crowd behaviour, based on visual information, and to add it to the other models based on information extracted by means of computer vision techniques, in a knowledge base which is supported by an inference engine.

Objectives

  • Critically review and analyse previous approaches to the field of crowd modelling and analysis in video surveillance systems
  • Development of a new crowd model based on the fusion of agent-based and entity-based approaches
  • Development of a feature extraction methods from video adapted to macroscopic and microscopic crowd level analyses
  • Development of a common-sense knowledge approach to crowd modelling
  • Development of a new architecture for hybrid knowledge representation which combines human common-sense knowledge and deep learning concepts
  • Development of a hybrid knowledge representation scheme and to integrate it with deep neural networks
  • Crowd behaviour recognition
  • Design a deep learning model for crowd scene analysis
  • Develop an experimental system and to test and evaluate it

Our Publications

Journals:

  • 1. Maleš, Lada ; Marčetić, Darijan ; Ribarić, Slobodan A multi-agent dynamic system for robust multi-face tracking, Expert systems with applications. 126 February (2019); 246-264 Link
  • 2. Cheng, Haitao ; Yan, Li ; Ma, Zongmin ; Ribarić, Slobodan Fuzzy spatio-temporal ontologies and formal construction based on fuzzy Petri nets, Computational Intelligence. 35 (2019), 1; 1-36 Link

Proceedings articles:

  • 1. Marčetić, Darijan; Ribarić, Slobodan; An Online Multi-Face Tracker for Unconstrained Videos, Proceedings SITIS (2018); 159-165 Link
  • 2. Marčetić, Darijan ; Maleš, Lada ; Ribarić, Slobodan Crowd Motion Pattern Detection at the Microscopic Level, Proceedings of the 42nd International Convention MIPRO 2019 May (2019); 1283-1288 Link
  • 3. Marčetić, Darijan; Ribarić, Slobodan. A Fuzzy Logic-based Approach to Detection of Abnormal Crowd Behaviour, Proceedings ELMAR-2019, 143-146 Link
  • 3. Matković, Franjo; Marčetić, Darijan; Ribarić, Slobodan. Abnormal Crowd Behaviour Recognition in Surveillance Videos, Proceedings SITIS-2019, 428-435 Link Presentation

Research student's projects:

Bachelor thesis:

Staff

TitleNamePhoneOfficeEmailPosition
Prof. dr. sc.Slobodan Ribarić +385 (01) 6129-952D-311slobodan.ribaric@fer.hrFull Professor / Chair of the Laboratory
Izv. prof. dr. sc.Tomislav Hrkać +385 (01) 6129-926D-344tomislav.hrkac@fer.hrResearcher
Dr. sc.Darijan Marčetić +385 (01) 6129-550D-336darijan.marcetic@fer.hrPostDoc Researcher
Izv. prof. dr. sc.Vitomir Štruc +386 (01) 4768-839 - vitomir.struc@fe.uni-lj.siResearcher
Dr. sc.Lada Maleš - - lmales@ffst.hrResearcher
Prof. dr. sc.Ivo Ipšić +385 (51) 584-709 - ivo.ipsic@riteh.hrResearcher
Dr. sc.Marina Ivašić-Kos +385 (51) 584-710 - marinai@uniri.hrResearcher
Mr. sc.Darko Mihalić +385 (01) 6129-550D-336darkox3@gmail.comPHD student
Dipl. ing.Marijo Maračić +385 (01) 6129-550D-336marijo.maracic@fer.hrPHD student
Mag. ing.Franjo Matković +385 (01) 6129-550D-336franjo.matkovic@fer.hrPHD student

Other projects

Some of the current and previous projects (both national and international) the Laboratory was involved in are given in the list below:

Copyright © 2006 Sitename.com
Designed by Web Page Templates
 
Design downloaded from free website templates.