Login (DCU Staff Only)
Login (DCU Staff Only)

DORAS | DCU Research Repository

Explore open access research and scholarly works from DCU

Advanced Search

Predicting media memorability using ensemble models

Azcona, David orcid logoORCID: 0000-0003-3693-7906, Moreu, Enric, Hu, Feiyan orcid logoORCID: 0000-0001-7451-6438, Ward, Tomás E. orcid logoORCID: 0000-0002-6173-6607 and Smeaton, Alan F. orcid logoORCID: 0000-0003-1028-8389 (2020) Predicting media memorability using ensemble models. In: MediaEval 2019, 27 - 29 Oct 2019, Sophia Antipolis, France.

Abstract
Memorability, defined as the quality of being worth remembering, is a pressing issue in media as we struggle to organize and retrieve digital content and make it more useful in our daily lives. The Predicting Media Memorability task in MediaEval 2019 tackles this problem by creating a challenge to automatically predict memorability scores building on the work developed in 2018. Our team ensembled transfer learning approaches with video captions using embeddings and our own pre-computed features which outperformed Medieval 2018’s state-of-the-art architectures.
Metadata
Item Type:Conference or Workshop Item (Paper)
Event Type:Workshop
Refereed:No
Subjects:Computer Science > Machine learning
DCU Faculties and Centres:DCU Faculties and Schools > Faculty of Engineering and Computing > School of Computing
Research Initiatives and Centres > INSIGHT Centre for Data Analytics
Published in: Proceedings of MediaEval 2019. . CEUR Workshop Proceedings.
Publisher:CEUR Workshop Proceedings
Official URL:http://ceur-ws.org/Vol-2670/
Copyright Information:© 2019 The Authors. CC-BY-4.0
Use License:This item is licensed under a Creative Commons Attribution-NonCommercial-Share Alike 3.0 License. View License
Funders:Science Foundation Ireland SFI/12/RC/2289
ID Code:23833
Deposited On:25 Oct 2019 11:54 by David Azcona . Last Modified 23 Sep 2020 10:38
Documents

Full text available as:

[thumbnail of David_Azcona_MediaEval_2019_Memorability_Camera_ready.pdf]
Preview
PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
2MB
Downloads

Downloads

Downloads per month over past year

Archive Staff Only: edit this record