ITI-CERTH participation in ActEV and AVS Tracks of TRECVID 2021

Konstantinos Gkountakos, Damianos Galanopoulos, Δέσποινα Τούσκα, Konstantinos Ioannidis, Stefanos Vrochidis, Vasileios Mezaris, Ioannis Kompatsiaris, Despoina TouskaZenodo 2022 · DOI · OpenAlex

Abstract

In this report, the overview of the runs during the TRECVID 2021 by the ITI-CERTH team are presented. ITI-CERTH participated in the Ad-hoc Video Search (AVS) and Activities in Extended Video (ActEV) tasks. For the AVS task, our participation is based on an attention-based cross-modal deep network architecture. As part of training this architecture, we experimented with a new hard negative<br> mining approach. For the ActEV task, we improve our framework, in terms of more accurate performance, by addressing the classification problem as multi-label rather than a single-label.
BibTeX
@inproceedings{gkountakos2022iti,
  author = {Konstantinos Gkountakos and Damianos Galanopoulos and Δέσποινα Τούσκα and Konstantinos Ioannidis and Stefanos Vrochidis and Vasileios Mezaris and Ioannis Kompatsiaris and Despoina Touska},
  title = {{ITI-CERTH participation in ActEV and AVS Tracks of TRECVID 2021}},
  booktitle = {Zenodo},
  year = {2022},
  doi = {10.5281/zenodo.5817708},
}

Loading related content...

Figure