Fig. 12

3D model of a forest environment in the lab. We used a mix of 20 small laurel and bay trees of under 2 m height to recreate a forest environment in the motion capture lab. The 3D model was captured using SemanticPaint and a Kinect v1 as a mapping sensor. Note that in this case the world coordinate system of the captured map is not aligned with that of the motion capture coordinate system