1 code implementation • 26 Mar 2024 • Saptarshi Sinha, Alexandros Stergiou, Dima Damen
We propose an exemplar-based approach that discovers visual correspondence of video exemplars across repetitions within target videos.
Ranked #1 on Repetitive Action Counting on RepCount (using extra training data)
1 code implementation • 3 Nov 2023 • Alexandros Stergiou, Brent De Weerdt, Nikos Deligiannis
We encode temporally occluded trajectories, jointly learn latent representations of the occluded segments, and reconstruct trajectories based on expected motions across different temporal segments.
Ranked #1 on Video Anomaly Detection on UBnormal
1 code implementation • ICCV 2023 • Alexandros Stergiou, Nikos Deligiannis
The success of deep learning models has led to their adaptation and adoption by prominent video understanding methods.
1 code implementation • 20 Oct 2022 • Alexandros Stergiou, Dima Damen
A key function of auditory cognition is the association of characteristic sounds with their corresponding semantics over time.
Ranked #4 on Audio Classification on EPIC-KITCHENS-100
1 code implementation • CVPR 2023 • Alexandros Stergiou, Dima Damen
We propose a bottleneck-based attention model that captures the evolution of the action, through progressive sampling over fine-to-coarse scales.
Ranked #1 on Early Action Prediction on UCF101
1 code implementation • 1 Nov 2021 • Alexandros Stergiou, Ronald Poppe
We evaluate adaUnPool on image and video super-resolution and frame interpolation.
no code implementations • 5 Oct 2021 • Alexandros Stergiou
The hierarchical extraction of features models variations of relatively similar classes the same as very dissimilar classes.
1 code implementation • 29 Jan 2021 • Alexandros Stergiou
Visual interpretability of Convolutional Neural Networks (CNNs) has gained significant popularity because of the great challenges that CNN complexity imposes to understanding their inner workings.
3 code implementations • ICCV 2021 • Alexandros Stergiou, Ronald Poppe, Grigorios Kalliatakis
Convolutional Neural Networks (CNNs) use pooling to decrease the size of activation maps.
1 code implementation • 8 Nov 2020 • Alexandros Stergiou, Ronald Poppe
To address this challenge, we present a novel spatio-temporal convolution block that is capable of extracting spatio-temporal patterns at multiple temporal resolutions.
Action Recognition In Videos Temporal Action Localization +1
1 code implementation • 15 Jun 2020 • Alexandros Stergiou, Ronald Poppe
Generalizing over temporal variations is a prerequisite for effective action recognition in videos.
Ranked #3 on Action Recognition on HACS
no code implementations • 7 Feb 2020 • Alexandros Stergiou, Ronald Poppe, Remco C. Veltkamp
We show that using Class Regularization blocks in state-of-the-art CNN architectures for action recognition leads to systematic improvement gains of 1. 8%, 1. 2% and 1. 4% on the Kinetics, UCF-101 and HMDB-51 datasets, respectively.
no code implementations • 30 Sep 2019 • Alexandros Stergiou, Ronald Poppe
Motivated by the often distinctive temporal characteristics of actions in either horizontal or vertical direction, we introduce a novel convolution block for CNN architectures with video input.
1 code implementation • 18 Sep 2019 • Alexandros Stergiou, Georgios Kapidis, Grigorios Kalliatakis, Christos Chrysoulas, Ronald Poppe, Remco Veltkamp
We demonstrate the method on six state-of-the-art 3D convolution neural networks (CNNs) on three action recognition (Kinetics-400, UCF-101, and HMDB-51) and two egocentric action recognition datasets (EPIC-Kitchens and EGTEA Gaze+).
1 code implementation • 4 Feb 2019 • Alexandros Stergiou, Georgios Kapidis, Grigorios Kalliatakis, Christos Chrysoulas, Remco Veltkamp, Ronald Poppe
Deep learning approaches have been established as the main methodology for video classification and recognition.
1 code implementation • 31 Jul 2018 • Alexandros Stergiou, Ronald Poppe
The main challenges stem from dealing with the considerable variation in recording setting, the appearance of the people depicted and the coordinated performance of their interaction.