Sökning: onr:"swepub:oai:DiVA.org:ltu-94540" >
VidHarm: A Clip Bas...
VidHarm: A Clip Based Dataset for Harmful Content Detection
-
- Edstedt, Johan (författare)
- Linköpings universitet,Datorseende,Tekniska fakulteten
-
- Berg, Amanda (författare)
- Linköpings universitet,Datorseende,Tekniska fakulteten
-
- Felsberg, Michael (författare)
- Linköpings universitet,Datorseende,Tekniska fakulteten
-
visa fler...
-
- Karlsson, Johan (författare)
- Statens Medieråd, Stockholm,Statens Medierad, Sweden
-
- Benavente, Francisca (författare)
- Statens Medieråd, Stockholm,Statens Medierad, Sweden
-
- Novak, Anette (författare)
- Statens Medieråd, Stockholm,Statens Medierad, Sweden
-
- Grund Pihlgren, Gustav, 1994- (författare)
- Luleå tekniska universitet,EISLAB,Lulea Univ Technol, Sweden
-
visa färre...
-
(creator_code:org_t)
- Institute of Electrical and Electronics Engineers (IEEE), 2022
- 2022
- Engelska.
-
Ingår i: 2022 26th International Conference on Pattern Recognition (ICPR). - : Institute of Electrical and Electronics Engineers (IEEE). - 9781665490627 - 9781665490634 ; , s. 1543-1549
- Relaterad länk:
-
https://urn.kb.se/re...
-
visa fler...
-
https://doi.org/10.1...
-
https://urn.kb.se/re...
-
visa färre...
Abstract
Ämnesord
Stäng
- 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
- NATURVETENSKAP -- Data- och informationsvetenskap -- Datorseende och robotik (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Computer Vision and Robotics (hsv//eng)
- NATURVETENSKAP -- Data- och informationsvetenskap -- Datavetenskap (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Computer Sciences (hsv//eng)
Nyckelord
- Visualization
- Annotations
- Pattern recognition
- Task analysis
- Age Rating
- Video
- Audio
- Maskininlärning
- Machine Learning
Publikations- och innehållstyp
- ref (ämneskategori)
- kon (ämneskategori)
Hitta via bibliotek
Till lärosätets databas