Song, Yang and Goncalves, Luis and Perona, Pietro (2003) Unsupervised learning of human motion. IEEE Transactions on Pattern Analysis and Machine Intelligence, 25 (7). pp. 814-827. ISSN 0162-8828 http://resolver.caltech.edu/CaltechAUTHORS:SONieeetpami03
See Usage Policy.
Use this Persistent URL to link to this item: http://resolver.caltech.edu/CaltechAUTHORS:SONieeetpami03
An unsupervised learning algorithm that can obtain a probabilistic model of an object composed of a collection of parts (a moving human body in our examples) automatically from unlabeled training data is presented. The training data include both useful "foreground" features as well as features that arise from irrelevant background clutter - the correspondence between parts and detected features is unknown. The joint probability density function of the parts is represented by a mixture of decomposable triangulated graphs which allow for fast detection. To learn the model structure as well as model parameters, an EM-like algorithm is developed where the labeling of the data (part assignments) is treated as hidden variables. The unsupervised learning technique is not limited to decomposable triangulated graphs. The efficiency and effectiveness of our algorithm is demonstrated by applying it to generate models of human motion automatically from unlabeled image sequences, and testing the learned models on a variety of sequences.
|Additional Information:||© Copyright 2003 IEEE. Reprinted with permission. Manuscript received 27 Sept. 2002; revised 2 Mar. 2003; accepted 17 Mar. 2003. [Posted online: 2003-06-20] Recommended for acceptance by V. Pavlovic. Part of the work in this paper was published in the Proceedings of IEEE Conference on Computer Vision and Pattern Recognition ’01 and NIPS ’01. This work was funded by the US National Science Foundation Engineering Research Center for Neuromorphic Systems Engineering (CNSE) at Caltech (NSF9402726), and an Office of Navy Research grant N00014-01-1-0890 under the MURI program. The authors would like to thank Charless Fowlkes for bringing the Chow and Liu paper to their attention.|
|Subject Keywords:||Unsupervised learning, human motion, decomposable triangulated graph, probabilistic models, greedy search, EM algorithm, mixture models|
|Usage Policy:||No commercial reproduction, distribution, display or performance rights in this work are provided.|
|Deposited By:||Archive Administrator|
|Deposited On:||11 Dec 2006|
|Last Modified:||26 Dec 2012 09:21|
Repository Staff Only: item control page