/KinectMocap4Blender

A Kinect motion capture plugin for Blender

Primary LanguagePythonMIT LicenseMIT

KinectMocap4Blender

A Kinect motion capture plugin for Blender

Presentation

This Blender add-on allows you to capture live movement for a human-like model in Blender, using a Microsoft Kinect V2 sensor. Although not precise, the result makes a good working base to reproduce real life movement after refining it manually. It is a cheap solution for homemade motion capture.

The target model must be in a standing rest pose.

It has been developped for Blender 2.79 Windows 10. There is also a version for Blender 2.80 but I advise strongly against using it (it has to stay in an experimental state until Blender 2.80 is officially out of Beta).

A few words about the project genesis

A few monthes ago, I browsed the web in order to find a way to setup my own homemade capture studio with a Kinect sensor, for a videogame I'm working on. I am not a 3D artist and a real noob with Blender, so I wanted something to help me with the animation process. I found a few solutions but they were either very expensive (and not working very well anyway, at least the demos I tried) or not fitting my needs (not working on Windows 10 or requiring a specific armature that just wouldn't do with my target model). So I decided to learn Python, dig into Blender documentation (and I think I broke a few shovels in the process ;)), and here is the result.

Demo and configuration video

[https://youtu.be/Zt8gJzSNSbw]

Install from release archive

  • Install Kinect for Windows SDK 2.0 if you haven't already
  • Download the release archive and unpack the files (kinecp_mocap.py and kinectMocap4Blender.pyd) corresponding to your version of Blender in Blender addons directory. Consult Blender documentation for more information on plugin installation. [https://docs.blender.org/]

Dependencies

Build

There are two parts in the project :

  • kinect_mocap.py : the blender add-on (different for Blender 2.79 and 2.80)
  • kinectMocap4Blender.pyd : a C++ library The library was designed to be built using Visual Studio 2017 (only the Release configuration has been properly configured at the moment).

Current progress

The project is currently in version 1.1.