Abstract:Depth sensor based 3D human motion estimation hardware such as Kinect has made interactive applications more popular recently. However, it is still challenging to accurately recognize postures from a single depth camera due to the inherently noisy data derived from depth images and self-occluding action performed by the user. In this paper, we propose a new real-time probabilistic framework to enhance the accuracy of live captured postures that belong to one of the action classes in the database. We adopt the Gaussian Process model as a prior to leverage the position data obtained from Kinect and marker-based motion capture system. We also incorporate a temporal consistency term into the optimization framework to constrain the velocity variations between successive frames. To ensure that the reconstructed posture resembles the accurate parts of the observed posture, we embed a set of joint reliability measurements into the optimization framework. A major drawback of Gaussian Process is its cubic learning complexity when dealing with a large database due to the inverse of a covariance matrix. To solve the problem, we propose a new method based on a local mixture of Gaussian Processes, in which Gaussian Processes are defined in local regions of the state space. Due to the significantly decreased sample size in each local Gaussian Process, the learning time is greatly reduced. At the same time, the prediction speed is enhanced as the weighted mean prediction for a given sample is determined by the nearby local models only. Our system also allows incrementally updating a specific local Gaussian Process in real time, which enhances the likelihood of adapting to run-time postures that are different from those in the database. Experimental results demonstrate that our system can generate high quality postures even under severe self-occlusion situations, which is beneficial for real-time applications such as motion-based gaming and sport training.

Model-based golf swing reconstruction

Smart Motion Reconstruction System for Golf Swing: a DBN Model Based Transportable, Non-Intrusive and Inexpensive Golf Swing Capture and Reconstruction System.

Dynamic Bayesian Network Model Based Golf Swing 3D Reconstruction Using Simple Depth Imaging Device

Three-Dimensional Human Modeling and Dynamics Simulation Based on ADAMS

Optimal State Estimation of Spinning Ping-Pong Ball Using Continuous Motion Model

Dynamic Human Body Reconstruction and Motion Tracking with Low-Cost Depth Cameras

Rebound Modeling of Spinning Ping-Pong Ball Based on Multiple Visual Measurements

Model-driven sketch reconstruction with structure-oriented retrieval

Model Based Motion State Estimation and Trajectory Prediction of Spinning Ball for Ping-Pong Robots Using Expectation-Maximization Algorithm

Vision based ping-pong ball tracking with physical model

Reconstructing 3D Human Models with a Kinect

Three-Dimensional Spatiotemporal Reconstruction and Feature Analysis of Table Tennis Movement Enhanced by Multi-view Computer Vision

GolfDB: A Video Database for Golf Swing Sequencing

Badminton Movement Quality Assessment Through Kinect Sensor Data Using NDT-GCN Algorithm

Knee joint model reconstruction and kinematic measurements using a statistical deformation model

A Semantic Web-Based Approach for Bat Trajectory Reconstruction With Human Keypoint Information

3D Kinematics Estimation from Video with a Biomechanical Model and Synthetic Training Data

Kinect Posture Reconstruction Based on a Local Mixture of Gaussian Process Models

An Assessment Method for Upper Limb Rehabilitation Training Using Kinect

Applying Pose Estimation to Predict Amateur Golf Swing Performance Using Edge Processing