Abstract:

Multimedia archives increasingly consist of data from multiple sources with different characteristics that can be exploited. In this talk we will focus on video classification over multiple-source collections, and discuss whether classifiers should train from individual sources or from the full data set. If training separately, how can we merge rank lists from different sources 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 perform significantly better than merging methods without explicitly learning the mapping functions.