| Autor: |
Gkountakos, Konstantinos, Galanopoulos, Damianos, Touska, Despoina, Ioannidis, Konstantinos, Vrochidis, Stefanos, Mezaris, Vasileios, Kompatsiaris, Ioannis |
| Jazyk: |
angličtina |
| Rok vydání: |
2023 |
| Předmět: |
|
| DOI: |
10.5281/zenodo.7431865 |
| Popis: |
This report presents the overview of the runs related to Ad-hoc Video Search (AVS) and Activitiesin Extended Video (ActEV) tasks on behalf of the ITI-CERTH team. Our participation in the AVStask is based on a cross-modal deep network architecture utilizing several textual and visual features.As part of the retrieval stage, a dual-softmax approach is utilized to revise the calculated text-video similarities. For the ActEV task, we adapt our framework to fit the new dataset and overcome thechallenges of detecting and recognizing activities in a multi-label manner while experimenting withtwo separate activity classifiers. |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
