Me

Alexandros Stergiou

Postdoc Researcher at Vrije Universiteit Brussel
alexandros.stergiou@vub.be

Short bio

My main research focus is on Computer Vision and Machine Learning

I am a Postdoc Researcher at Vrije Universiteit Brussel (VUB) working on explainability methods for video models. I was previously a Research Associate at the University of Bristol working with Dima Damen on video understanding. I obtained my PhD from Utrecht University's Department of Information and Computing Science where I was lucky to be supervised by Ronald Poppe and Remco C. Veltkamp. The topic of my thesis was based on human action and interaction recognition in everyday social settings. During my time in Utrecht, I also worked on improving the efficiency and explainability of spatiotemporal deep learning models, for scene understanding and classification of videos. I completed my Master's and Bachelor's degrees in Computer Science at the University of Essex.




PUBLICATIONS

PUBLICATIONS LIST

A full list can be found at my GoogleScholar profile.


ARXIV PREPRINTS

leaps leaps-active

Stergiou A., and Deligiannis N.,Leaping Into Memories: Space-Time Deep Feature Synthesis
ArXiv

CONFERENCE PAPERS

tempr tempr-active

Stergiou A., and Damen D., The Wisdom of Crowds: Temporal Progressive Attention for Early Action Prediction
to be presented in CVPR 2023
playitback playitback-active

Stergiou A., and Damen D., Play It Back: Iterative Attention for Audio Recognition
to be presented in ICASSP 2023
mindseye mindseye-active

Stergiou A., Poppe R., and Kalliatakis G., Refining activation downsampling with SoftPool
ICCV 2021
mindseye mindseye-active

Stergiou A., The Mind's Eye: Visualizing Class Agnostic Features of CNNs,
ICIP 2021
mtconv mtconv_active

Stergiou A., and Poppe R., Multi-Temporal Convolutions for Human ActionRecognition in Videos
IJCNN 2021

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
ICCVw 2019
FAST_conv

Stergiou A., and Poppe R., Spatio-Temporal FAST 3D Convolutions for Human Action Recognition
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
ICIP 2019

JOURNAL PAPERS

mindseye mindseye-active

Stergiou A., and Poppe R., AdaPool: Exponential Adaptive Pooling for Information-Retaining Downsampling
IEEE Transactions on Image Processing, 2023
learn2cycle srtg-active

Stergiou A., and Poppe R., Learn to cycle: Time-consistent feature discovery for action recognition
Pattern Recognition Letters, 2021
h2h_survey

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

DATASETS

Fig1
Inter4K



THESIS

thesis-static thesis-active

Stergiou A.G., Efficient Modelling Across Time of Human Actions and Interactions
Utrecht University 2021

EXPERIENCE

PAST AND CURRENT ROLES

aff1
October 2022 - now
Postdoctoral Researcher at Vrije Universiteit Brussel, Brussels, BE
PI: Nikos Deligiannis

aff1
August 2021 - September 2022
Research Associate at University of Bristol, Bristol, UK
PI: Dima Damen

EDUCATION

aff1
September 2017 - September 2021
PhD student at Utrecht University, Utrecht, NL
Thesis title: Efficient Modelling Across Time of Human Actions and Interactions
Supervisor: Ronald Poppe and Remco C. Veltkamp
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: Adrian Clark
aff2
September 2013 - September 2016
BSc student in Computer Science at University of Essex, Colchester, UK




ACTIVITIES

TEACHING

Fig1
Lecturer for INFOMCV (Masters course) Spring 2018, 2019 & 2020
Lecturer and Instructor for the Deep Learning/Neural Networks lectures

REVIEWING

Journals: TPAMI, IJCV, TIP, Pattern Recognition
Conferences: ECCV 2022, CVPR 2022-2023
Outstanding Reviewer: ECCV 2022

© Alexandros Stergiou and Deborah Dike