Collision detection for volume sample? #54
Description
Currently collision detection is performed using the complete bounding box of the volume irrespective of whether the voxel are transparent or not.
Can the bounding box be computed to include only the visible voxels?
Can a structure such as Bounding volume hierarchy be used to get better collision detection accuracy?