MLVMA2011 Technical Program
08:50-09:00
Opening Address
09:00-10:00 Invited Talk 1 : Learning feature hierarchies for image and video analysis
Prof. Yann LeCun, New
York University,
USA
10:30-11:10
Session 1: Pose and Motion Estimation
Kernel
PLS regression for robust monocular pose estimation
Radu
Dondera and Larry Davis, University of Maryland,
USA
Joint
gait-pose manifold for video-based human motion estimation
Xin
Zhang and Guoliang Fan, Oklahoma State University, USA
11:10
- 12:10 Session 2: Object Tracking
(*) Tracking through scattered
occlusion
Haggai
Abramson and Shai Avidan, Tel Aviv
University, Israel
Non-rigid
tracking of musk shrews in video for detection of emetic
episodes
Dong
Huang, Carnegie Mellon University, USA
Kelly Meyers, Sverine Henry, Fernando De la
Torre, and Charles Horn, University of Pittsburgh,
USA
Occlusion
robust multi-camera face tracking
Josh
Harguess, Changbo Hu, and J. K. Aggarwal, University of Texas at Austin,
USA
12:10
- 14:00 Lunch Break
14:00-15:00 Invited Talk 2 : Sparse Learning Methods for Image and Video Analysis
Prof.
Dimitris
Metaxas,
Rutgers University, USA
15:00
- 15:30 Coffee Break
15:30-16:10:
Session 3: Action Recognition
Human
action recognition in crowded surveillance video sequences by using features
taken from key-point trajectories
Masaki
Takahashi, Japan Broadcasting Corporation, Japan
Mahito Fujii, Masahide Naemura, and Shinichi
Satoh, National Institute of Informatics,
Japan
HMM-MIO:
an enhanced hidden Markov model for action recognition
Massimo
Piccardi, Richard Yi Da Xu, Oscar Perez Concha, and Zia Moghaddam, University
of Technique, Sydney, Australia
16:10
- 16:50 Session 4: Motion Pattern Analysis
Improved
anomaly detection in crowded scene via cell-based analysis of foreground speed,
size and textures
Vikas
Reddy, Conrad Sanderson, and brain Lovell, NICTA,
Australia
(**) Dynamic modeling of streaklines for
motion pattern analysis in video
Nandita
Nayak and Amit Roy-Chowdhury, University of California, Riverside, USA
Note: The two talks denoted by (*) and (**) will be switched.