[x3d-public] General Progress

[Joseph D Williams]

We may be a bit short of some actual current news about web3d x3d hanim but for all, let it be known that V2.0 now showing at web3d.org includes more complete hands and feet and is still looking to have the annex A example humanoid user code updated to the latest and greatest joint, segment, and site names and hierarchy for the 5 levels of articulation. Since the basic hanim is solid, this is not a big deal but, for me, just waiting to see that all is settled before finally getting the examples updated. 

Those that might have wished for unfiltered medtech naming of objects will not be disappointed because many hands and feet names have been deemed more medically accurate and complete and thus are different if only in name. In return, V2.0 has gained in its documentary character by providing more accurate and complete illustrations of the various articulation hierarchies. 

Please notice that the V2.0 has extended itself to deal with the acquisition, conversion, normalization, and playback of data obtained from realtime motion capture systems. Since the basic idea is to prepare a playback skeleton to execute the data sampled from your capture skeleton, if the models match initial pose and articulation, then maybe just a matter of converting the possibly BVH-style data to the ‘standard’ x3d axis-angle and getting it into appropriate timers and interpolators. Then you can start to have some real fun because now you have control. 

The big hints: 1. capture skeleton bone orientation is the same as hanim playback skeleton joint rotation as long as you are picking the correct playback joint; 2. Use the hanim playback skeleton initial pose as the initial pose for your capture model; 3a. Choose the highest articulation playback skeleton from the beginning because then you easily can add animations of joints not captured; 3b. Edit timesensor intervals and joint axis-angle pitch, yaw, and roll axes to control the animations; 3c. See the similarities between axis-angle form and the quaternion form because the quats are really the current animation transport standard; and 4. If you really want to know how to mocap surface details then animate vertices of playback skin, then it is the same but just a bit different. 

The web3d conference phone meetings are Wednesdays at 5pm pst. I think participants have refined definitions in some minor areas, added some new opportunities for improvements in details for actuation of the model, and have complete web3D Consortium support for this actual World Standard used to construct realistic ‘standard’ humanoids executing realistic ‘standard’ animations. 
 
Thanks and Best,  
Joe
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://web3d.org/pipermail/x3d-public_web3d.org/attachments/20181014/d21d055c/attachment.html>