Man

Sergio Alejandro Velastin

Visiting Professor

Queen Mary University of London

London, Reino Unido

Líneas de Investigación


Computer Vision, Deep Learning, Object localisation and classification, Intelligent Transport Systems

Educación

  •  BSc Honours, University of Manchester. Reino Unido, 1978
  •  MSc, University of Manchester. Chile, 1979
  •  PhD, University of Manchester. Chile, 1982

Experiencia Académica

  •   Senior Research Fellow Part-time

    Queen Mary University of London

    London, Reino Unido

    2014 - Sin Información

  •   Professor Full-time

    Universidad de Santiago de Chile

    Santiago, Chile

    2013 - Sin Información

  •   Catedratico de Excelencia Full-time

    Universidad Carlos III de Madrid

    Madrid, España

    2013 - 2014

  •   Professor/Director Digital Imaging Research Centre Full-time

    Kingston University

    London, Reino Unido

    2001 - 2013

  •   Senior Lecturer Full-time

    Kings College London

    London, Reino Unido

    2001 - 2001

  •   Marie Curie Research Professor Full Time

    UNIVERSIDAD CARLOS III DE MADRID

    Engineering

    Madrid, España

    2015 - 2018

Experiencia Profesional

  •   Founder/Reseach Director/Consultant

    Ipsotek Ltd.

    London, Reino Unido

    2001 - Sin Información

  •   Marie Curie Research Professor Full Time

    Universidad Carlos III de Madrid

    Madrid, España

    2015 - 2018

  •   Senior Researcher Full Time

    Cortexica Vision Systems Ltd

    London, Reino Unido

    2018 - 2019

Formación de Capital Humano


Research supervision (PhD), all completed, Kings College London to 2001: “Mahalanobis Distance Hough Transform with Extended Kalman Filtering”, “Transputer-based computer architectures for cooperating robots”, “Crowd monitoring using image processing”, “Towards verification of MASCOT designs using Petri Nets”, “Coloured Petri Nets for software design verification”, “Modelling and Verification of MASCOT designs using LOTOS”, “Real-time motion estimation for complex image scenes”, “Pedestrian detection, tracking and behaviour classification”.

Research Student Supervision (PhD) Kingston University 2001-2012: “Tracking people in crowds across multiple cameras” (Damien Simmonet, completd 2014), “Identification of unusual behaviour in crowds” (Beibei Zhan, completed 2008), “Distributed Architecture for Visual Surveillance Systems” (Valera, Completed 2007), “Multi camera tracking” (F Yin, completed), “Visual surveillance for road traffic monitoring” (Buch, Completed), “Tag and Track in multi-camera networks” (Colombo, completed), "An Architectural Framework for Intelligent Spaces" (R. Herrera, completed subject to minor revisions, 2014)



 

Article (59)

Spatial deep feature augmentation technique for FER using genetic algorithm
Train Station Pedestrian Monitoring Pilot Study Using an Artificial Intelligence Approach
3D Object Detection for Self-Driving Cars Using Video and LiDAR: An Ablation Study
BERT for Activity Recognition Using Sequences of Skeleton Features and Data Augmentation with GAN
Experimental Method to Estimate the Density of Passengers on Urban Railway Platforms
Camera and LiDAR analysis for 3D object detection in foggy weather conditions
Human Activity Recognition by Sequences of Skeleton Features
Human Fall Detection using YOLO: A Real-Time and AI-on-the-Edge Perspective
Lightweight CNN and GRU Network for Real-Time Action Recognition
Video augmentation technique for human action recognition using genetic algorithm
3D Object Detection with SLS-Fusion Network in Foggy Weather Conditions
Facial Expression Recognition of Instructor Using Deep Features and Extreme Learning Machine
Fall Detection and Activity Recognition Using Human Skeleton Features
Fall detection using human skeleton features
Feature Selection Using Correlation Analysis and Principal Component Analysis for Accurate Breast Cancer Diagnosis
Multiple object tracking for robust quantitative analysis of passenger motion while boarding and alighting a metropolitan Train
T-VLAD: Temporal vector of locally aggregated descriptor for multiview human action recognition
A unified deep framework for joint 3D pose estimation and action recognition from a single RGB camera
Agglomerative Clustering and Residual-VLAD Encoding for Human Action Recognition
Breast Tumor Classification Using an Ensemble Machine Learning Method
Detecting, Tracking and Counting People Getting On/Off a Metropolitan Train Using a Standard Video Camera
Multiple Batches of Motion History Images (MB-MHIs) for Multi-view Human Action Recognition
Vectors of temporally correlated snippets for temporal action detection
Vehicle Make and Model Recognition Using Bag of Expressions
Bag of Deep Features for Instructor Activity Recognition in Lecture Room
Detection and Tracking of Motorcycles in Congested Urban Environments Using Deep Learning and Markov Decision Processes
Dynamic Spatio-Temporal Bag of Expressions (D-STBoE) Model for Human Action Recognition
End-to-End Temporal Action Detection Using Bag of Discriminant Snippets
Instructor Activity Recognition through Deep Spatiotemporal Features and Feedforward Extreme Learning Machines
Learning to recognise 3D human action from a new skeleton-based representation using deep convolutional neural networks
Spatio-Temporal Image Representation of 3D Skeletal Movements for View-Invariant Action Recognition with Deep Convolutional Neural Networks
TAB: Temporally aggregated bag-of-discriminant-words for temporal action proposals
3D-HOG EMBEDDING FRAMEWORKS FOR SINGLE AND MULTI-VIEWPOINTS ACTION RECOGNITION BASED ON HUMAN SILHOUETTES
A Bag of Expression framework for improved human action recognition
DA-VLAD: DISCRIMINATIVE ACTION VECTOR OF LOCALLY AGGREGATED DESCRIPTORS FOR ACTION RECOGNITION
Evaluating a bag-of-visual features approach using spatio-temporal features for action recognition
Exploiting deep residual networks for human action recognition from skeletal data
Feature Similarity and Frequency-Based Weighted Visual Words Codebook Learning Scheme for Human Action Recognition
PMHI: Proposals From Motion History Images for Temporal Segmentation of Long Uncut Videos
SKELETAL MOVEMENT TO COLOR MAP: A NOVEL REPRESENTATION FOR 3D ACTION RECOGNITION WITH INCEPTION RESIDUAL NETWORKS
People detection and pose classification inside a moving train using computer vision
Shadow Detection for Vehicle Classification in Urban Environments
Multi-view human action recognition using 2D motion templates based on MHIs and their HOG description
Vision-based traffic surveys in urban environments
A Kinect-based 3D hand-gesture interface for 3D databases
Multi-view Human Action Recognition using Histograms of Oriented Gradients (HOG) Description of Motion History Images (MHIs)
An optimisation of Gaussian mixture models for integer processing units
Learning multi-planar scene models in multi-camera videos
Local feature saliency classifier for real-time intrusion monitoring
Structural Laplacian Eigenmaps for Modeling Sets of Multivariate Sequences
3D Interaction environment for free view point TV and games using multiple tablet computers
An approach for protecting transport infrastructure
Urban road user detection and classification using 3D wire frame models
Crowd analysis: a survey
Editorial: Visual information engineering
Modelling periodic scene elements for visual surveillance
Time efficient ghost removal for motion detection in visual surveillance systems
Novel concepts and challenges for the next generation of video surveillance systems
Calibration and object correspondence in camera networks with widely separated overlapping views

ConferencePaper (2)

F1 score assesment of Gaussian mixture background subtraction algorithms using the MuHAVi dataset
Evaluation of background subtraction algorithms using MuHAVi, a multicamera human action video dataset

Editorial (1)

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics): Preface

Proyecto (38)

Catedra de Excelencia
LASIE=> Large Scale Information Exploitation of Forensic Data
OBSERVE
Autonomous Pedestrian Detection (principal investigator=> Dr Pablo Zegers)
ADDPRIV (Automatic Data relevancy Discrimination for a PRIVacy-sensitive video surveillance)
Evaluation of Appearance-based Pedestrian Detection
HAR, "Human Action Recognition"
Surveillance Video Data Mining
A DVR for VCA (Video Contents Analysis)
An Implementation of a MIPSA Surveillance System for Rome
PROTECTRAIL (Protection of European Railways from Security and Safety Threats)
Tag and Track for Security and Forensics
Tracking individuals in crowds with a mixed network of static and PTZ (Pan-Tilt-Zoom) cameras
Experimental Study of Vehicle Passenger Interactions at Public Transport Stations (PI=> Prof. Rodrigo Fernandez, I was internallional research collaborator)
A UK-Japan Partnership on Ambient Intelligence
FARM, "Framework for the Analysis of Rich Media"
Robust wide-area multi-camera tracking of people and vehicles to improve CCTV visualisation",
CARETAKER=> “Content Analysis and REtrieval Technologies to Apply Knowledge Extraction to massive Recordings”
Visual Modelling of People Behaviours and Interactions for Professional Skills Training”, European Office of Aerospace Research
Robust Methods for Monitoring and Understanding People in Public Spaces
Surveillance, Evaluation, Research, Validation, and Exploitation
Getting the best use of CCTV in the Railways
COmputational HEteRogEneously timed NeTworks
ADVISOR=> Annotated Digital Video for Surveillance and Optimised Retrieval
PRISMATICA=> “Pro-active Integrated Systems for Security Management by Technological Institutional and Communication Assistance”
Assessment of Image Processing techniques as a means of improving Personal Security in Public Transport
CROMATICA=> “Crowd Management with Telematic Imaging and Communications Assistance”
Automatic Data relevancy Discrimination for a Privacy-Sensitive Video Surveillance
Autonomous Pedestrian Detection
Catedra de Excelencia
Evaluation of Appearance-based Pedestrian Detection
Experimental Study of Vehicle Passenger Interactions at Public Transport Stations
Framework for the Analysis of Rich Media
Protection of European Railways from Security and Safety Threats
Surveillance Video Data Mining
Tag and Track for Security and Forensics
Tracking individuals in crowds with a mixed network of static and PTZ (Pan-Tilt-Zoom) cameras
View-independent Human Action Recognition from Single Uncalibrated Camera

Review (1)

Detection of Motorcycles in Urban Traffic Using Video Analysis: A Review
74
Sergio Velastin

Visiting Professor

School of Elec. Eng. and Computer Science

Queen Mary University of London

London, Reino Unido

2
Sebastián Seriani

Profesor Jerarquizado

Pontificia Universidad Católica de Valparaíso

Valparaíso, Chile

2
Raúl Herrera

Instructor A

UNIVERSIDAD DE TARAPACA

Chile

1
Rodrigo Fernández

Profesor Titular

Obras Civiles

UNIVERSIDAD DE LOS ANDES

Santiago, Chile

1
Gonzalo Garcia

Postdoc

College of Engineering

Virginia Commonwealth University

Richmond, Estados Unidos