Browse DORAS
Browse Theses
Search
Latest Additions
Creative Commons License
Except where otherwise noted, content on this site is licensed for use under a:

Empirical exploration of extreme SVM-RBF parameter values for visual object classification

Albatal, Rami and Little, Suzanne (2014) Empirical exploration of extreme SVM-RBF parameter values for visual object classification. In: MMM 2014, The 20th Anniversary International Conference on MultiMedia Modeling, 6-10 Jan 2014, Dublin, Ireland. ISBN DOI: 10.1007%2F978-3-319-04117-9_28

Full text available as:

[img]PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
364Kb

Abstract

This paper presents a preliminary exploration showing the surprising effect of extreme parameter values used by Support Vector Machine (SVM) classifiers for identifying objects in images. The Radial Basis Function (RBF) kernel used with SVM classifiers is considered to be a state-of-the-art approach in visual object classification. Standard tuning approaches apply a relative narrow window of values when determining the main parameters for kernel size. We evaluated the effect of setting an extremely small kernel size and discovered that, contrary to expectations, in the context of visual object classification for some object and feature combinations these small kernels can demonstrate good classification performance. The evaluation is based on experiments on the TRECVid 2013 Semantic INdexing (SIN) training dataset and provides initial indications that can be used to better understand the optimisation of RBF kernel parameters.

Item Type:Conference or Workshop Item (Paper)
Event Type:Conference
Refereed:Yes
Uncontrolled Keywords:Visual Object Classification; SVM; RBF; Optimisation; Extreme parameter values
Subjects:Computer Science > Machine learning
Computer Science > Artificial intelligence
Computer Science > Digital video
DCU Faculties and Centres:Research Initiatives and Centres > INSIGHT Centre for Data Analytics
Published in:Multimedia Modelling. Lecure Notes in Computer Science 8326. Springer. ISBN DOI: 10.1007%2F978-3-319-04117-9_28
Publisher:Springer
Use License:This item is licensed under a Creative Commons Attribution-NonCommercial-Share Alike 3.0 License. View License
Funders:European Framework Programme 7
ID Code:19592
Deposited On:21 Jan 2014 11:17 by Rami Albatal. Last Modified 21 Jan 2014 11:17

Download statistics

Archive Staff Only: edit this record