Estimation of motions and tracking the motions with the help of a motion tracking system has been cons idered dilemma due to the availability of various framewor ks. In the process of tracking, we need to form a m apping from observation space to the state space. We have revie wed Gaussian Process with both discriminative and generative motion estimation. Both the input and output are sh own at the same place. But there are certain limitations due to the dilemma of multi modality. To overcome these limitations we have used a mix of Gaussian process experts. In this paper, we have c ombined both these information's into a unified system, whi ch acts as a fusion of experts. We have used Gaussian process, dynamic model for learning the movements in latent state space. We had not only estimated human motions but also tracked motions with respe ct to other objects. The area of study discussed in this paper would be use ful to trace out physical and behavioral patterns o f living and non living object and this would help in studying the forecast patterns such as natural blooming, industr y tool des earthquakes forecast , monitoring and in control a pplications along with their impact so as to offer corrective Estimation of motions and tracking the motions with the help of a motion tracking system has been cons idered as a dilemma due to the availability of various framewor ks. In the process of tracking, we need to form a m apping from observation space to the state space. We have revie wed Gaussian Process with both discriminative and generative motion estimation. Both the input and output are sh own at the same place. But there are certain limitations due to the dilemma of multi modality. To overcome these limitations we have used a mix of both these information's into a unified system, whi ch acts as a fusion of experts. We have used Gaussian process, dynamic model for learning the movements in latent ct to other objects. The area of study discussed in this paper would be use ful to trace out physical and behavioral patterns o f living and non living object and this would help in studying the forecast patterns such as natural blooming, industr y tool des ign or earthquakes forecast , monitoring and in control a pplications along with their impact so as to offer corrective
[1]
David J. Fleet,et al.
Gaussian Process Dynamical Models
,
2005,
NIPS.
[2]
Carl E. Rasmussen,et al.
Infinite Mixtures of Gaussian Process Experts
,
2001,
NIPS.
[3]
Zoubin Ghahramani,et al.
Sparse Gaussian Processes using Pseudo-inputs
,
2005,
NIPS.
[4]
Yun Fu,et al.
Human Pose Regression Through Multiview Visual Fusion
,
2010,
IEEE Transactions on Circuits and Systems for Video Technology.
[5]
David J. Fleet,et al.
3D People Tracking with Gaussian Process Dynamical Models
,
2006,
2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06).
[6]
Neil D. Lawrence,et al.
Fast Sparse Gaussian Process Methods: The Informative Vector Machine
,
2002,
NIPS.
[7]
A. Elgammal,et al.
Inferring 3D body pose from silhouettes using activity manifold learning
,
2004,
CVPR 2004.
[8]
Yun Fu,et al.
Temporal-Spatial Local Gaussian Process Experts for Human Pose Estimation
,
2009,
ACCV.
[9]
Ankur Agarwal,et al.
Recovering 3D human pose from monocular images
,
2006,
IEEE Transactions on Pattern Analysis and Machine Intelligence.
[10]
Carl E. Rasmussen,et al.
A Unifying View of Sparse Approximate Gaussian Process Regression
,
2005,
J. Mach. Learn. Res..
[11]
David J. Fleet,et al.
Priors for people tracking from small training sets
,
2005,
Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.