Multimedia Event Detection Using Hidden Conditional Random Fields
作者： Kimiaki Shirahama , Marcin Grzegorzek , Kuniaki Uehara
关键词: Intermediate layer 、 Computer science 、 Event structure 、 TRECVID 、 Machine learning 、 Classifier (UML) 、 Artificial intelligence 、 Multimedia 、 Conditional random field
摘要: This paper introduces a method for Multimedia Event Detection (MED). Given training videos certain event, classifier is constructed to identify displaying it. In particular, the problems of weakly supervised setting and unclear event structure are addressed in this paper. The first issue associated with loosely annotated that usually contain many irrelevant shots. second one difficulty assuming advance, because can be created by arbitrary camera editing techniques. To overcome these problems, Hidden Conditional Random Field (HCRF) used where hidden states work as an intermediate layer discriminate between relevant shots event. addition, relation among characterises structure. Thus, above managed optimising their relation, so distinguish occurs from rest videos. Experimental results on TRECVID video data validate effectiveness HCRFs context.
acm.org
sci-hub.se 参考文章(25)
Jingchen Liu, Scott McCloskey, Yanxi Liu, Local Expert Forest of Score Fusion for Video Event Classification Computer Vision – ECCV 2012. pp. 397- 410 ,(2012) , 10.1007/978-3-642-33715-4_29
Yun Zhai, Zeeshan Rasheed, Mubarak Shah, A Framework for Semantic Classification of Scenes Using Finite State Machines conference on image and video retrieval. pp. 279- 288 ,(2004) , 10.1007/978-3-540-27814-6_35
Zaid Harchaoui, Matthijs Douze, Dan Oneata, Relja Arandjelovic, Jérôme Revaud, Tinne Tuytelaars, Robin Aly, Omkar M. Parkhi, Danila Potapov, Cordelia Schmid, Kevin McGuinness, Noel E. O'Connor, Jakob Verbeek, Ken Chatfield, Shu Chen, Basura Fernando, Andrew Zisserman, Heng Wang, Jochen Schwenninger, AXES at TRECVid 2012: KIS, INS, and MED Proceedings TRECVid 2012. pp. 1- 13 ,(2012)
Qiang Yang, Derek Hao Hu, Jie Yin, Spatio-temporal event detection using dynamic conditional random fields international joint conference on artificial intelligence. ,vol. 2009, pp. 1321- 1326 ,(2009)
Rehan Akbani, Stephen Kwek, Nathalie Japkowicz, Applying support vector machines to imbalanced datasets european conference on machine learning. ,vol. 3201, pp. 39- 50 ,(2004) , 10.1007/978-3-540-30115-8_7
Georges Quénot, Stéphane Ayache, Video corpus annotation using active learning european conference on information retrieval. pp. 187- 198 ,(2008) , 10.5555/1793274.1793300
Mark D. Smucker, James Allan, Ben Carterette, A comparison of statistical significance tests for information retrieval evaluation conference on information and knowledge management. pp. 623- 632 ,(2007) , 10.1145/1321440.1321528
Jianguo Zhang, Shaogang Gong, Action categorization with modified hidden conditional random field Pattern Recognition. ,vol. 43, pp. 197- 203 ,(2010) , 10.1016/J.PATCOG.2009.05.015
Kimiaki Shirahama, Kuniaki Uehara, A Novel Topic Extraction Method Based on Bursts in Video Streams multimedia and ubiquitous engineering. pp. 249- 252 ,(2008) , 10.1109/MUE.2008.101
Ryoichi Ando, Koichi Shinoda, Sadaoki Furui, Takahiro Mochizuki, Robust scene recognition using language models for scene contexts multimedia information retrieval. pp. 99- 106 ,(2006) , 10.1145/1178677.1178693