360 likes | 716 Views
3D Human Body Pose Estimation using GP-LVM. Moin Nabi Computer Vision Group Institute for Research in Fundamental Sciences (IPM). Introduction to Human Pose Estimation. Articulated pose estimation from single-view monocular image(s). Application of Human Pose Estimation.
E N D
3D Human Body Pose Estimation using GP-LVM Moin Nabi Computer Vision Group Institute for Research in Fundamental Sciences (IPM)
Introduction to Human Pose Estimation Articulated pose estimation from single-view monocular image(s)
Application of Human Pose Estimation ■ Entertainment: Animation, Games ■ Security: Surveillance ■ Understanding: Gesture/Activity recognition
Difficulties of Human Pose estimation ■ Appearance/size/shape of people can vary dramatically ■ The bones and joints are observable indirectly (obstructed by clothing) ■ Occlusions ■ High dimensionality of the state space ■ Lose of depth information in 2D image projections
Difficulties of Human Pose estimation ■Challenging Human Motion
Problem Backgrounds ■ Pose Estimation From Monocular Image Goal: Reliable 3D Human Pose Estimation from single-camera input
Gaussian process a 5x5 covariance matrix and a 3-d input vector was used to calculate the 2-d output mean vector and the corresponding variances
Gaussian process Use for Regression
Linear Dimension Reduction Find the best latent inputs by maximizing the marginal likelihood under the constraint that all visible variables must share the same latent values.
Linear Dimension Reduction Find the best latent inputs by maximizing the marginal likelihood under the constraint that all visible variables must share the same latent values.
Human Pose Estimationusing GP-LVM Image -> Pose In Latent Space
Human Pose Estimationusing GP-LVM Motion capture example, representing 102-D data in 2-D
Pose from Action Thank You
Future Work Different Action has Different shape in latent space Guess Action from shape of model in latent space