Trademark matching and retrieval in sports video databases.
Lists
Only you see this
Abstract
In this paper we describe a system for detection and retrieval of trademarks appearing in sports videos. We propose a compact representation of trademarks and video frame content based on SIFT feature points. This representation can be used to robustly detect, localize, and retrieve trademarks as they appear in a variety of different sports video types. Classification of trademarks is performed by matching a set of SIFT feature descriptors for each trademark instance against the set of SIFT features detected in each frame of the video. Localization is performed through robust clustering of matched feature points in the video frame. Experimental results are provided, along with an analysis of the precision and recall. Results show that the our proposed technique is efficient and effectively detects and classifies trademarks.
BibTeX
@inproceedings{bagdanov2007trademark,
author = {Andrew D. Bagdanov and Lamberto Ballan and Marco Bertini and Alberto Del Bimbo},
title = {{Trademark matching and retrieval in sports video databases.}},
booktitle = {Multimedia Information Retrieval},
year = {2007},
doi = {10.1145/1290082.1290096},
}Loading related content...