Output details
11 - Computer Science and Informatics
University of Westminster
Article title
Active foreground region extraction and tracking for sports video annotation
Type
D - Journal article
Title of journal
Neural Processing Letters
Article number
-
Volume number
37
Issue number
1
First page of article
33
ISSN of journal
1370-4621
Year of publication
2013
Number of additional authors
3
Additional information
<23> Originality: This paper proposes a new framework for identifying dominant active regions within videos. The framework considers the distribution of entropy within a set of video frames, to cluster image features and identify regions of interest.
Significance: The algorithm addresses an important aspect of analysis of existing videos as it eliminates the need for cameral calibration and any a priori information about the active regions or the background.
Rigour: Extensive experimentation through benchmark videos. Results published in a peer-reviewed journal.
Interdisciplinary
-
Cross-referral requested
-
Research group
None
Citation count
0
Proposed double-weighted
No
Double-weighted statement
-
Reserve for a double-weighted output
No
Non-English
No
English abstract
-
