Title :
Merging rank lists from multiple sources in video classification
Author :
Lin, Wei-Hao ; Hauptmann, Alexander
Author_Institution :
Sch. of Comput. Sci., Carnegie Mellon Univ., Pittsburgh, PA
Abstract :
Multimedia corpora increasingly consist of data from multiple sources, with different characteristics that can be exploited by specialized applications. This paper focuses on video classification over multiple-source collections, and addresses the question whether classifiers should train from individual sources or from a full data set across all sources. If training separately, how can rank lists from different sources be merged effectively? We formulate the problem of merging ranked lists as learning a function mapping from local scores to global scores, and propose a learning method based on logistic regression. In our experiments we find that source characteristics are very important for video classification. Moreover, our method of learning mapping functions performs significantly better than merging methods without explicitly learning the mapping junctions
Keywords :
content-based retrieval; image classification; merging; multimedia databases; relevance feedback; video databases; function mapping; global scores; learning method; local scores; logistic regression; multimedia corpora; multiple-source collections; rank list merging; video classification; Application software; Computer science; Contracts; Learning systems; Logistics; Merging; Round robin; Statistics;
Conference_Titel :
Multimedia and Expo, 2004. ICME '04. 2004 IEEE International Conference on
Conference_Location :
Taipei
Print_ISBN :
0-7803-8603-5
DOI :
10.1109/ICME.2004.1394539