Crowd analysis using Deep Neural Networks
Principal Supervisor: Dr. Jesus Martinez del Rincon
+ Project Description
Crowd control and crow monitoring has become a necessity to preserve public safety due to our current life style, concentrated in highly populated cities and urban areas. This has become more evident after recent terrorist attacks such as Boston 2015 or Nice 2016. Manual motoring has proved inefficient due to the fact that human being suffer to perform even the simplest task such as crowd counting or anomaly detection in high density images. Better analysis and understanding of crowd behaviour through automatic and efficient intelligent systems in our cities is therefore a societal need. While initial attempts have been made in the literature, automatic analysis of highly congested and crowded scenes is one of the most challenging vision tasks that remain to be solved.
The aim of this work is to address the problem of crowd analysis and understanding in video sequence by using deep neural networks. Deep learning architectures have been shown to achieve state-of-art performance in multiple image recognition and object detection tasks, and they have recently provided preliminary results on simple and isolated crowd analysis task, such as person counting [1-4]. By building on this research, this project proposes a holistic view of the crow analysis problem through the use or multi-task and unsupervised learning to develop novel deep architectures able to deal with the complexity of the congested scenes. The undergoing research also aims to provide the candidate with state-of-art knowledge for the highly demanded data scientist jobs in the market.
, Chuan Wang, Hua Zhang, Liang Yang, Si Liu, Xiaochun Cao (2015) Deep People Counting in Extremely Dense Crowds, in Proceedings of the 23rd ACM international conference on Multimedia MM’15, pp 1299-1302
 Cong Zhang, Hongsheng Li, Xiaogang Wang, Xiaokang Yang (2015) Cross-scene Crowd Counting via Deep Convolutional Neural Networks, in Computer Vision and Pattern Recognition CVPR 2015
 Lokesh Boominathan, Srinivas S S Kruthiventi, R. Venkatesh Babu (2016)CrowdNet: A Deep Convolutional Network for Dense Crowd Counting, n Proceedings of the 23rd ACM international conference on Multimedia MM’16
 Mark Marsden, Kevin McGuinness, Suzanne Little, Noel E. O’Connor (2017)ResnetCrowd: A Residual Deep Learning Architecture for Crowd Counting, Violent Behaviour Detection and Crowd Density Level Classification, arXiv preprint arXiv:1705.10698
• To apply convolutional neural networks on raw video data and optical flow input information to extract features able to distinguish and characterise crowds and their individual.
• To investigate and develop new deep architectures able to solve such as crowd counting and crowd density estimation.
• To explore multi task learning and multi-branch deep networks able to fully analyse and recognise crowd behaviours.
• To investigate unsupervised deep learning paradigms to detect anomalous and potentially violent individual behaviours within the crowd
+ Academic Requirements
Students entering the programme will normally be required to have a 2.1 BSc/BEng in Computer Science, Electrical and Electronic Engineering, or a maths based engineering or physical science degree, or equivalent qualification recognised by the University. Students holding an appropriate MEng or MSc (Software conversion) will normally be required to have a 2.1 or commendation (distinction) respectively. Furthermore, additional criteria may be applied. All applicants must have significant mathematical and programming experience.
+ How to Apply And General Information
This 4 year PhD studentship, potentially funded by the Department for Employment and Learning (DEL), commences on 1 October 2018.
Eligibility for both fees and maintenance depends on the applicants being either an ordinary UK resident or those EU residents who have lived permanently in the UK for the 3 years immediately preceding the start of the studentship. Non UK residents who hold EU residency may also apply but if successful may receive fees only.
Applicants should apply electronically through the Queen’s online application portal at: https://dap.qub.ac.uk/portal/
Further information available at: http://www.qub.ac.uk/schools/eeecs/StudyattheSchool/PhDProgrammes
+ Contact Details
|Supervisor Name:||Dr. Jesus Martinez del Rincon|
+44 (0)28 9097 1779