Semantic Concept Discovery for Large-Scale Zero-Shot Event Detection
Xiaojun Chang, Yi Ping Yang, Alexander G. Hauptmann, Eric P. Xing, Yaoliang Yu · 2015
We focus on detecting complex events in uncon-strained Internet videos. While most existing works rely on the abundance of labeled training data, we consider a more difficult zero-shot setting where no training data is supplied. We first pre-train a number of concept classifiers using data from other sources. Then we evaluate the semantic correla-tion of each concept w.r.t. the event of interest. Af-ter further refinement to take prediction inaccuracy and discriminative power into account, we apply the discovered concept classifiers on all test videos and obtain multiple score vectors. These distinct score vectors are converted into pairwise compari-son matrices and the nuclear norm rank aggregation framework is adopted to seek consensus. To ad-dress the challenging optimization formulation, we propose an efficient, highly scalable algorithm that is an order of magnitude faster than existing alter-natives. Experiments on recent TRECVID datasets verify the superiority of the proposed approach. 1