In addition to “Position” constraints, there are a number of other types of constraints, that provide alternative ways for the markers to be driven by their target.
They are all assigned the same way: select the motion capture markers, then the bone it is to drive, and click “pos”, “ori”, “both”, “slide” or “aim”. Markers will be parented to the skeleton, and connected to the source object. The source does not have to be motion capture markers. Any object that has a world position in Maya can be used.
The name of the marker does not need to be the same, nor does the source marker need to be in a particular hierarchy. The relationship is defined by creating a connection of the “worldMatrix” attribute.
If there is only one selection, it will set the selection to being an active marker of that type.
 | Match on global position (world translation) |
 | Match on global orientation (world rotation) |
 | Match on global position and orientation (world translation and rotation) |
 | Match on position, but allow marker to slide along axis of the child transform. |
 | Make marker point towards target, but don’t pull. |
When attaching motion capture data to a skeleton, the “pos” constraint is generally the most applicable. Slide may also be useful for markers located around the elbows and knees.
If the entity that is driving the transform, also contains the rotation information, for instance, a rigid body made up of three markers, the “ori” or “both” constraints can be used. For instance, if three or four markers for the head are turned in to one marker that has rotation, then that marker can be used to drive the head joint with a “both” or “ori” constraint.
The ability to solve based on both translation and rotation allows for solving for a wide range of setups rather than just point cloud data on to a skeleton hierarchy. It is possible to solve from skeleton to skeleton, or create alternative character rigs that go beyond the traditional motion capture setups.
Tip: When using “Both” or “Ori” markers, the orientation target is solved at the position of the marker. In many cases you want the orientation to be applied at the joint. For this reason it’s generally best to remove any translation on the target marker (active marker). Also, its best to avoid using “Both” markers; in many cases using both a “pos” and an “ori” marker is more effective, as the translation on the ori marker can be removed so the rotation is solved without any offset.