THUMOS'13: International Workshop on Action Recognition with a Large Number of Classes In conjunction with ICCV'13, December 7, 2013, Sydney, Australia CHALLENGE RESULTS ARE AVAILABLE NOW: http://crcv.ucf.edu/ICCV13-Action-Workshop/ Automatically recognizing and localizing a large number of action categories from videos "in the wild" is of significant importance for video understanding and multimedia event detection. THUMOS aims at exploring new challenges and approaches for large-scale action recognition with large number of classes from open source videos. The workshop program which includes invited talks and presentations of the challenge top performers can be found below: Opening Session 09:00 Opening Remarks, Rahul Sukthankar, Google Research 09:05 Invited Talk and Challenge Top Performer Presentation: Cordelia Schmid (INRIA), "Actions Recognition From Videos: Some Recent Results" 09:35 Invited Talk: Jason J. Corso (SUNY at Buffalo), "Analyzing the Role of Motion and Time in Large-scale Action Recognition" 10:05 Coffee Break Action Classification Session 10:30 Overview and the Results of the Classification Challenge, Jingen Liu, SRI International 10:50 Challenge Top Performer: "L1-regularized Logistic Regression Stacking and CRF Smoothing for Action Recognition", S. Karaman, L. Seidenari, A. Bagdanov, A. Del Bimbo 11:10 Challenge Top Performer: "Hybrid Super Vector with Improved Dense Trajectory Features", X. Peng, L. Wang, Z. Cai, Y. Qiao 11:30 Invited Talk: Silvio Savarese (Stanford), "Recognizing Human Actions in Context" 12:00 Challenge Top Performer: "Ordered Trajectories for Large Scale Human Action Recognition", R. Oruganti, R. Goecke 12:20 "A Spatio-Temporal Feature based on Triangulation of Dense SURF", K. Yanai, D. Nga 12:40 Lunch Action Localization Session 14:30 Overview and the Results of the Localization Challenge, Yu-Gang Jiang, Fudan University 14:50 Invited Talk and Challenge Top Performer Presentation: Jianxin Wu (Nanjing University), "From Image to Video Encoding for Action Recognition" 15:20 Coffee Break Conclusion Session 16:10 Invited Talk: Tal Hassner (Open University, Israel), "What do action recognition benchmarks tell us about action recognition capabilities?" 16:40 Summary of the Challenge Results and Closing Remarks, Massimo Piccardi, Univ. of Tech., Sydney 17:00 End More information available at the workshop website: http://crcv.ucf.edu/ICCV13-Action-Workshop/ Challenge results and summary of the submissions: http://crcv.ucf.edu/ICCV13-Action-Workshop/results.html Program Chairs: Jingen Liu, SRI International Yu-Gang Jiang, Fudan University Amir Roshan Zamir, UCF General Chairs: Ivan Laptev, INRIA Massimo Piccardi, Univ. of Tech., Sydney Massimo Piccardi, UCF Rahul Sukthankar, Google Research