Full-body movement tracking with Kinect and Neuron Mocap

Full-body movement tracking with Kinect and Neuron Mocap.

Neuron Perception Log

Description:

Perception neuron is a small, adaptive, and affordable motion capture technology base on IMU (Inertial Measurement Unit) composed of a 3-axis Gyroscope, 3-axis Accelerometer and 3-axis Magnetometer. Perception neuron delivers smooth true motion with minimal latency to VR enthusiasts. However, it still has the problem of drift. In this project, we are going to introduce an absolute sensor, e.g. Microsoft Kinect, to eliminate the drift from IMU and correct joint rotation. In a simple case, by doing a reaching task using perception neuron suit, palm rotation becomes unreliable due to the drift issue. But Microsoft Kinect SDK provides a reliable palm rotation that could be used with IMU data. Data fusion, measurement and compensation will be done in Unity3D game engine. Game development with Untiy3D/ C# and data analysis are required in this project.

 

 

Achievements:

Minimum achievements:

  • Recording a movement with 3 motion capture systems: Phasespace, Kinect and Perception neuron should be conducted, e.g. reaching task with upper limb.
  • Data analysis with system Kinect and Perception neuron should be done, comparing with the ground truth data from Phasespace.
  • Find a solution to fuse Kinect data with IMU data to minimize the drift effect.
  • Compare the Kinect+IMU with Phasespace data.

 

Advanced achievements:

  • Game design with Kinect + IMU mocap system.

 

Advanced achievements:

  • Game design with Kinect + IMU mocap system.

 

Information, materials and resource:

Untiy3D game engine: http://unity3d.com/learn

Phasespace active tracking system: www.phasespace.com/

Perception Neuron: https://neuronmocap.com/

Some paper to start with:

Paper Upper limb motion tracking with the integration of IMU and Kinect 

https://hal.inria.fr/lirmm-00623141/document

www.mdpi.com/1424-8220/15/8/20097/pdf

 

 

Contact:

Nan. WANG@epfl.ch

Ronan.BOULIC@epfl.ch