Sökning: onr:"swepub:oai:DiVA.org:ltu-94540" >
VidHarm: A Clip Bas...
-
Edstedt, JohanLinköpings universitet,Datorseende,Tekniska fakulteten
(författare)
VidHarm: A Clip Based Dataset for Harmful Content Detection
- Artikel/kapitelEngelska2022
Förlag, utgivningsår, omfång ...
-
Institute of Electrical and Electronics Engineers (IEEE),2022
-
printrdacarrier
Nummerbeteckningar
-
LIBRIS-ID:oai:DiVA.org:ltu-94540
-
https://urn.kb.se/resolve?urn=urn:nbn:se:ltu:diva-94540URI
-
https://doi.org/10.1109/ICPR56361.2022.9956148DOI
-
https://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-191876URI
Kompletterande språkuppgifter
-
Språk:engelska
-
Sammanfattning på:engelska
Ingår i deldatabas
Klassifikation
-
Ämneskategori:ref swepub-contenttype
-
Ämneskategori:kon swepub-publicationtype
Anmärkningar
-
ISBN för värdpublikation: 978-1-6654-9062-7
-
Funding Agencies|ELLIIT; Strategic Area for ICT research - Swedish Government; Vinnova [2020-04057]; Wallenberg Artificial Intelligence, Autonomous Systems and Software Program (WASP) - Knut and Alice Wallenberg Foundation
-
Automatically identifying harmful content in video is an important task with a wide range of applications. However, there is a lack of professionally labeled open datasets available. In this work VidHarm, an open dataset of 3589 video clips from film trailers annotated by professionals, is presented. An analysis of the dataset is performed, revealing among other things the relation between clip and trailer level annotations. Audiovisual models are trained on the dataset and an in-depth study of modeling choices conducted. The results show that performance is greatly improved by combining the visual and audio modality, pre-training on large-scale video recognition datasets, and class balanced sampling. Lastly, biases of the trained models are investigated using discrimination probing.VidHarm is openly available, and further details are available at the webpage https://vidharm.github.io/
Ämnesord och genrebeteckningar
Biuppslag (personer, institutioner, konferenser, titlar ...)
-
Berg, AmandaLinköpings universitet,Datorseende,Tekniska fakulteten(Swepub:liu)amabe60
(författare)
-
Felsberg, MichaelLinköpings universitet,Datorseende,Tekniska fakulteten(Swepub:liu)micfe03
(författare)
-
Karlsson, JohanStatens Medieråd, Stockholm,Statens Medierad, Sweden
(författare)
-
Benavente, FranciscaStatens Medieråd, Stockholm,Statens Medierad, Sweden
(författare)
-
Novak, AnetteStatens Medieråd, Stockholm,Statens Medierad, Sweden
(författare)
-
Grund Pihlgren, Gustav,1994-Luleå tekniska universitet,EISLAB,Lulea Univ Technol, Sweden(Swepub:ltu)gusgru
(författare)
-
Linköpings universitetDatorseende
(creator_code:org_t)
Sammanhörande titlar
-
Ingår i:2022 26th International Conference on Pattern Recognition (ICPR): Institute of Electrical and Electronics Engineers (IEEE), s. 1543-154997816654906279781665490634
Internetlänk
Hitta via bibliotek
Till lärosätets databas