Video summarization based on Subclass Support Vector Data Description

Mygdalis, Vasileios; Iosifidis, Alexandros; Tefas, Anastasios; Pitas, Ioannis

doi:10.1109/cies.2014.7011849

Cited by 8 publications

(9 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, one-class classification methods show good performance when only the class of interest needs to be modeled and discriminated from the rest of the world. Moreover one-class approaches have been used in visual data classification problems, such as video surveillance and video summarization [103]. State-of-the-art one-class classification methods, such as the One-class Support Vector Machines (OC-SVM) [104], the Support Vector Data Description [105] and Least Squares One-Class Support Vector Machine (LS-OC-SVM) [106], achieve significantly better performance in their kernel version over their linear alternatives.…”

Section: Approximate Methods For Big Media Data Classificationmentioning

confidence: 99%

Big Data Analysis for Media Production

et al. 2016

Self Cite

View full text Add to dashboard Cite

Abstract-A typical high-end film production generates several terabytes of data per day, either as footage from multiple cameras or as background information regarding the set (laser scans, spherical captures, etc). This paper presents solutions to improve the integration of the multiple data sources, and understand their quality and content, which are useful both to support creative decisions on-set (or near it) and enhance the post-production process. The main cinema specific contributions, tested on a multisource production dataset made publicly available for research purposes, are the monitoring and quality assurance of multi-camera set-ups, multisource registration and acceleration of 3D reconstruction, anthropocentric visual analysis techniques for semantic content annotation, and integrated 2D-3D web visualization tools. We discuss as well improvements carried out in basic techniques for acceleration, clustering and visualization, which were necessary to deal with the very large multisource data, and can be applied to other big data problems in diverse application fields.

show abstract

Section: Approximate Methods For Big Media Data Classificationmentioning

confidence: 99%

Big Data Analysis for Media Production

et al. 2016

Self Cite

View full text Add to dashboard Cite

show abstract

“…Finally, the hyperpshere center is expressed as a linear combination of the determined support vectors. Applications of SVDD have been found in several fields, including hyperspectral image classification [2], [3], mechanical failure detection [4], biomedical data classification [5], face recognition [6], video summarization [7] and other outlier detection tasks [8], [9].…”

Section: Introductionmentioning

confidence: 99%

“…That is, they minimize the class variance within the SVDD optimization process. In [7], an additional clustering step is performed in the training data, in order to determine subclasses within the training class. The subclass information is thereby employed in the SVDD optimization process.…”

Section: Introductionmentioning

confidence: 99%

Exploiting local and global geometric data relationships in Support Vector Data Description

Mygdalis

Tefas

Pitas

2016

2016 23rd International Conference on Pattern Recognition (ICPR)

Self Cite

View full text Add to dashboard Cite

“…In order to efficiently model a class of interest in media classification tasks, we consider the use of One-Class Classification (OCC) methods [1,2,3,4,5]. Related OCC applications include hyperspectral image classification [3], video summarization [6,7], image segmentation [8]. Other OCC use case scenarios include applications when only one class is well sampled and must be distinguished from every other possibility, e.g., medical diagnostic problems, faults and failure detection, video surveillance and mobile fraud detection [9].…”

Section: Introductionmentioning

confidence: 99%