News

Invited talk to the PKM Series on Data Science for Business organised by MERCU BUANA University

Jan 19, 2021

I am very proud to have been invited as a speaker to the PKM Series on Data Science for Business organised by MERCU BUANA University, Jakarta, Indonesia. It was a pleasure to talk on Video Analysis for Business in this live stream event reaching primarily Indonesia, Malasya and the United Kingdom.

Latest research presented at the 5th FOLDOUT Consortium Meeting

Oct 5, 2020

I am very proud to have presented my latest research on person tracking and human behaviour analysis during the 5^th Consortium Meeting of the EU-FOLDOUT project. FOLDOUT focus is on through foliage detection of illegal activity at borders in the inner and outermost regions of the EU (http://www.foldout.eu ). Person tracking and human behaviour analysis can help in this context to detect abnormal human behaviour as well as suspicious activities at the borders.

Invited talk for the IET on Understanding Human behaviour from video

Oct 30, 2017

I was honoured to be invited by The Institution of Engineering and Technology (IET) to deliver a Lecture on Understanding Human behaviour from video at the University of Reading as part of their monthly seminars. The IET characterises for featuring first class speakers, hand-picked for their expertise and industry knowledge (www.theiet.org).

IEToct2017.pdf

People meet and Leave: Extracting human behaviour from video (Invited talk at Reading University )

Oct 20, 2016

I was invited to deliver a talk on human behaviour recognition within the series of seminars on Computer Science at University of Reading. The lecture took place on 19.10.2016.

Abstract:

Behaviour analysis is an essential task in modern video surveillance systems. Recently special attention has been placed on social or interaction detection between people in the monitored space. Detecting people meeting or splitting from groups is thus of highest interest to understand when a group is forming and potentially detecting a threat such as a fight or an attack to a person. In this talk I will present a novel approach to detect people meeting/splitting and potential threats. The proposed approach works by translating people behaviour from trajectory information into semantic terms. Having available a semantic model of the targeted behaviour, the event detection is performed in the semantic domain. The model is learnt employing a soft-computing clustering algorithm that combines trajectory information and motion semantic terms. I will present results obtained on a series of videos with different behaviour situations.

Live abnormal behaviour recognition performed in Sweden got noticed by the press

Oct 20, 2016 3:27:06 PM

I was in Linkoeping, Sweden, from 2nd to 6th October to perform a live demonstration on recognising abnormal behaviours on a monitored space by networked cameras. The demonstration was part of the final evaluation of the EU project P5 (www.p5-fp7.eu) by the European Commission. Abnormalities such as people loitering or detected in forbidden zones were detected with high accuracy. The work was noticed by the Swedish press. The article title reads 'Your behaviour can be monitored'.

Google Sites

Report abuse