Topics

• HBAxSCES is meant to tackle the topics mentioned above by looking at contributions falling within the following research areas:

• Image Analysis and Processing

Image Analysis and Image Processing allow extracting features relevant to Human Behaviour Understanding. Both low-level features such as texture, boundaries, textures, keypoints, and high-level features such as faces, objects, text, iris, body shapes, hands, and arms can be extracted to check on and analyse Human Behaviour within Smart City case scenarios.

• • Computer Vision

Computer Vision techniques are widely adopted to tackle topics such as motion, and trajectory extraction from video sequences as well as eye-tracking and human-computer interaction techniques that closely relate to Human Behaviour Understanding within Smart City Environments. The topics mentioned above are relevant to providing high-level information (suspicious behaviour alerts, usual and unusual events in crowds, real-time dangerous item detection).

• • Deep Learning and Machine Learning

Nowadays, several Human Behaviour Analysis methods are reliant on ever-increasing competitive artificial intelligence-based architectures. Current trends in scientific research see lots of novelties and outstanding accuracy rates on topics such as object detection, crowd tracking, people re-identification, face detection and recognition, sentiment analysis. Furthermore, it is observed an increasing interest in unsupervised and self-supervised deep learning architecture tackling the lack of labelled and manually annotated datasets in some of the above-mentioned topics.

• • Simulations

Simulation of behaviours in emergencies is an interesting subject that helps to understand evacuation processes and to give out warnings for contingency plans. Individual and crowd behaviours in the earthquake are different from those under normal circumstances. Panic will spread in the crowd and cause chaos. Without considering emotion, most existing behavioural simulation methods analyse the movement of people from the point of view of mechanics.

• • Digital Twins

The development of digital twins leads to analysing real-time situations and increasing the awareness and risks of potential physical attacks on urban critical infrastructure specifically. Digital twins play such a critical role in extracting knowledge, process and fusing observation data and information, prior to machine reasoning.