Me

Alexandros Stergiou

Utrecht University Computer Science Ph.D. student
a.g.stergiou@uu.nl

Bio

My work is focused on Computer Vision and Machine Learning

I am a PhD student at Utrecht University's Department of Information and Computing Sciences. My project is based on Machine Learning and Computer Vision for human action and interaction recognition in everyday social settings, supervised by dr. Ronald Poppe and prof. dr. Remco C. Veltkamp. During my time at Utrecht I have also worked on improvements on the efficiency and explainabilty of spatio-temporal deep learning models and methods, for scene understanding and classification of videos.
I obtained my Masters in Advance Computer Science from University of Essex working with dr. Adrian Clark. I have also worked shortly at the Institute for Analytics and Data Science (IADS) for a summer project on Neural Networks for structured population data with dr. Spyros Samothrakis.

PUBLICATIONS

PUBLICATIONS LIST

A full list can be found in my GoogleScholar profile.


mindseye mindseye-active

Stergiou A., Poppe R., and Kalliatakis G., Refining activation downsampling with SoftPool, arXiv:2101.00440, 2021

mindseye mindseye-active

Stergiou A., The Mind's Eye: Visualizing Class Agnostic Features of CNNs, International Confernce on Image Procesing (ICIP), 2021

mtconv mtconv_active

Stergiou A., and Poppe R., Multi-Temporal Convolutions for Human ActionRecognition in Videos, International Joint Conference of Neural Networks (IJCNN), 2021

learn2cycle srtg-active

Stergiou A., and Poppe R., Learn to cycle: Time-consistent feature discovery for action recognition, Pattern Recognition Letters, 2021

class-reg human-human-active

Stergiou A., Poppe R., and Veltkamp R.C., Learning Class-Specific Features with Class Regularization for Videos, Applied Sciences, 2020

feature-pyramids human-human-active

Stergiou A., Kapidis G., Kalliatakis G., Chrysoulas C., Poppe R., and Veltkamp R.C., Class Feature Pyramids for Video Explanation Intenrantional Conference on Computer Vision Workshop (ICCVW), 2019

FAST_conv

Stergiou A., and Poppe R., Spatio-Temporal FAST 3D Convolutions for Human Action Recognition, Intenrantional Conference on Machine Learning Applications (ICMLA), 2019

saliency-tubes human-human-active

Stergiou A., Kapidis G., Kalliatakis G., Chrysoulas C., Veltkamp R.C., and Poppe R., Saliency Tubes: Visual Explanations for Spatio-Temporal Convolutions, International Confernce on Image Procesing (ICIP), 2019

h2h_survey

Stergiou A., and Poppe R., Analyzing human-to-human interactions: a survey, Computer Vision and Image Understanding, 2019

TEACHING

MODULES THAT I HAVE BEEN INVOLVED IN

Fig1
INFOMCV (2019-2020) - Computer Vision (Master Course)
Lecturer for the Deep Learning lectures
Fig1
INFOMCV (2018-2019) - Computer Vision (Master Course)
Lecturer and Instructor for the Deep Learning/Neural Networks lectures
Fig1
INFOMCV (2017-2018) - Computer Vision (Master Course)
Instructor for the Deep Learning/Neural Networks lectures

EXPERIENCE

PRESENT & PAST EXPERINECE AND EDUCATION

aff1
August 2021 - now
Postdoctoral Research Associate at University of Bristol, Bristol, UK

aff1
September 2017 - September 2021
PhD student at Utrecht University, Utrecht, NL
Thesis title: Efficient Modelling Across Time of Human Actions and Interactions
Supervisor: prof. dr. Remco C. Veltkamp, Co-Supervisor: dr. Ronald Poppe
aff1
June 2016 - September 2017
Researcher at Institute of Analytics and Data Science (IADS), University of Essex, Colchester, UK
Supervisor: dr. Spyros Samothrakis
aff2
September 2016 - September 2017
MSc student in Advanced Computer Science at University of Essex, Colchester, UK
Dissertation title: The Driver's Assistant: Utilising Synthetic Data Generation and Deep Learning for Traffic Sign Classification
Supervisor: dr. Adian Clark
aff2
September 2013 - September 2016
BSc student in Computer Science at University of Essex, Colchester, UK




© Alexandros Stergiou