[x3d-public] What is "loa" ?

GPU Group gpugroup at gmail.com
Wed Jun 24 15:19:10 PDT 2020


In freewrl DEF/USE is just for the parser, for when you want to ROUTE or
re-USE a node. Once its parsed we don't refer to those DEF names in HAnim.
In freewrl, the Motion nodes -including HAnimMotion and (MotionPlay +
MotionClip) don't need the DEF names  Just the joint name="l_shoulder"
names.
-Doug

On Wed, Jun 24, 2020 at 4:03 PM Joseph D Williams <joedwil at earthlink.net>
wrote:

>
>    - Can we change the USE reference node to SFString instead of SFNode?
>
>
>
> That does nothing but mess up USE even more.
>
> The main idea is that the USE statements refer to the DEF name which does
> not match the joint names in the Motion node. Am I wrong about this? The
> DEF names are different from the name field  in the Joint. Sooner or later
> the Motion animation stuff has to know both. So what to do?
>
> It is obvious that the Motion node needs to know the Joint name first, to
> see if there are matches, then needs to find the DEF names to run the
> animation. What to do? I didn’t like or see the need for the Humanoid
> joints, segments, and sites fields because if xml, then it is easy to use
> DOM interfaces to find out the list of Joint nodes and give both the DEF
> name and the name name.
>
> Joe
>
>
>
>
>
>
>
> *From: *John Carlson <yottzumm at gmail.com>
> *Sent: *Wednesday, June 24, 2020 1:23 PM
> *To: *Joseph D Williams <joedwil at earthlink.net>
> *Cc: *Don Brutzman <brutzman at nps.edu>; GPU Group <gpugroup at gmail.com>; X3D
> Graphics public mailing list <x3d-public at web3d.org>
> *Subject: *Re: [x3d-public] What is "loa" ?
>
>
>
> Can we change the USE reference node to SFString instead of SFNode?
>
>
>
> On Wed, Jun 24, 2020 at 3:15 PM Joseph D Williams <joedwil at earthlink.net>
> wrote:
>
> Also if not too late, which it definitely should not be, Please change the
> Humanoid joints, segments, and sites fields to hold an MFString “string’
> ‘string’…” style that includes all Joint names, Segment names, and Site
> names. This aligns with the needs of the Humanoid Motion node that needs
> the actual names, not the DEFs.
>
> Including these in a USE instance is confusing because the DEF actually
> needs to be ignored by the hanim browser as far as rendering in concerned,
> and does not align with the needs of the Humanoid Motion node because that
> node needs the actual Joint names, not the actual Joint animation interface
> DEF names.
>
>
>
> As I said, I think I finally can agree with including those Humanoid
> fields if they actually serve a purpose. That purpose is to deliver sets of
> actual names that can be used by the Humanoid Motion node.
>
> Or, not a favorite of mine, the Humanoid Motion node should be changed to
> specify the Joint DEF names, which whatever is under the covers would have
> to discover anyway.
>
>
>
> Thanks and Best,
>
> Joe
>
>
>
>
>
> *From: *GPU Group <gpugroup at gmail.com>
> *Sent: *Wednesday, June 24, 2020 8:57 AM
> *To: *Don Brutzman <brutzman at nps.edu>; Joseph D Williams
> <joedwil at earthlink.net>
> *Cc: *X3D Graphics public mailing list <x3d-public at web3d.org>
> *Subject: *Re: [x3d-public] What is "loa" ?
>
>
>
> Don - if not too late for v4, facilitation and.or next steps requested for
> following topic:
>
> v4 HAnimMotion > MF field types in (web3d supplemental) MotionClip node
>
> Thanks,
>
> -Doug Sanden
>
> more..
>
> There may be a consensus forming around the idea of the extra nodes -
> MotionPlay and MotionClip - having MF fields for channels, joints (and
> values).. The 2 extra nodes aren't needed for full compliance to HAnim2 -
> they are supplemental - and so don't need to strictly comply. with field
> syntax.
>
> 1) HAnimMotion as specified in HANim2 with SF fields for channels, joints
> and values
>
> -- complies with HAnim2 specs fully
>
> 2) HAnimMotionPlay + HAnimMotionClip extra / supplemental nodes
>
> - MotionClip could have MF fields for channels, joints and values to
> assist authoring tool support
>
> - (in theory could have both SF and MF under different names if needed for
> abstract node inheritance ie parsedChannels MFString?)
>
> - and url field for loading .bvh files
>
> -- an SFNode nameMap field will help with some bvh use cases
>
>
>
> Related to
>
>
>
> http://dug9.users.sourceforge.net/web3d/tests/hanim/web3d/HAnim_Motion.pdf
>
>
>
> Mantis (resolved): 0001312: Mismatched type defaults defined for
> HAnimMotion channels field in X3D4
>
>
>
> On Tue, Jun 23, 2020 at 9:36 PM Joseph D Williams <joedwil at earthlink.net>
> wrote:
>
> Basically, me like very much
>
> Joe
>
>
>
> *From: *GPU Group <gpugroup at gmail.com>
> *Sent: *Monday, June 22, 2020 5:07 PM
> *To: *Joseph D Williams <joedwil at earthlink.net>
> *Cc: *X3D Graphics public mailing list <x3d-public at web3d.org>
> *Subject: *Re: [x3d-public] What is "loa" ?
>
>
>
> But I think there's room for both methods.
>
> - the Motion node probably needs to conform to HAnim 2.0 specifications
> already registered with (whatever body - is it ISO?).
>
> - a few more non-hanim-standard nodes could be considered
>
> - I proposed a few - HAnimMotionPaly + HAnimMotionData/HAnimMotionDataFile
>
> -- Don suggested merging HAnimMotionData and DataFile. I suggested
> HAnimMotionClip for the name
>
> - HAnimMotionClip has the 3 data fields from Motion. But when loading from
> url/,.bvh file, it doesn't populate those fields. Just uses them like
> HAnimMotion when no url given.
>
> So one way to make more of us happy:
>
> - keep the HAnimMotion node as defined for HAnim 2,0 specs
>
> - Do the MotionPlay + MotionClip additional nodes
>
> -- put a url field tor Doug (and the freewrlians who like the option to
> play with freely downloaded .bvh mocap files)
>
> -- put MF fields for Joe who likes style sheet conversions and authoring
> tool workflows
>
> How does that sound?
>
> -Doug
>
>
>
>
>
> On Mon, Jun 22, 2020 at 5:40 PM GPU Group <gpugroup at gmail.com> wrote:
>
> I can type in an image pixel by pixel into an MF field. Its more
> convenient to give a url to an image file.
>
> Same with .bvh.
>
> -Doug
>
>
>
> On Mon, Jun 22, 2020 at 5:34 PM Joseph D Williams <joedwil at earthlink.net>
> wrote:
>
>
>    - Or - have web3d browsers load .bvh directly, via a mapping / lookup
>    node / field./ table which you would likely also need for import tools.
>
>
>
> As discussed in another thread mainly about actually importing a complete
> bvh file, that is a legacy style that has the ‘official’ playback  skeleton
> is hidden (hanim V1 and x3d  exposed it using the Humanoid skeleton field)
> and the ‘user’ just imports his custom bvh file that is already matched to
> the internal hidden playback skeleton and then it gets executed and user
> gets some results. Maybe a video, even. Definitely not designed for
> realtime interactions.
>
>
>
> As is seen using Motion node, this style looks easy at first, then gets
> hard when you actually want to take control by editing the motion data, or
> to load another data set, or to update the ‘standard’ playback skeleton or
> imported capture skeleton.
>
>
>
> x3d hanim gets away from this legacy owner/user model by exposing the
> playback skeleton and the animation structures directly to the author. And,
> by using a coding scheme where animations do not need to be a part of the
> Humanoid user code we are more happy.
>
> Thanks,
>
> Joe
>
>
>
> *From: *Joseph D Williams <joedwil at earthlink.net>
> *Sent: *Monday, June 22, 2020 3:49 PM
> *To: *GPU Group <gpugroup at gmail.com>; X3D Graphics public mailing list
> <x3d-public at web3d.org>
> *Subject: *Re: [x3d-public] What is "loa" ?
>
>
>
>    - If so, then -as we get further from copy and pasta-ability- there
>    may be some demand for tools to import .bvh and convert to these x3d MF
>    fields.
>
>
>
> Right, I think the best idea would be to have a simple tool take the bvh
> file and turn it into xml of gltf style that can be validated and
> reformatted for editing. And, unless we want to treat strings like mf
> numbers, strings need delimiting quotes.
>
> Thanks for the hanim thoughts.
>
> Joe
>
>
>
>
>
> *From: *GPU Group <gpugroup at gmail.com>
> *Sent: *Saturday, June 20, 2020 1:23 PM
> *To: *X3D Graphics public mailing list <x3d-public at web3d.org>
> *Subject: *Re: [x3d-public] What is "loa" ?
>
>
>
> Hypothesis: the hanim motion fields were originally designed to make it
> easy to copy and paste from freely available motion capture .bvh files.
>
> If so, then -as we get further from copy and pasta-ability- there may be
> some demand for tools to import .bvh and convert to these x3d MF fields.
>
> Or - have web3d browsers load .bvh directly, via a mapping / lookup node /
> field./ table which you would likely also need for import tools.
>
> -Doug
>
>
>
> On Sat, Jun 20, 2020 at 12:38 PM GPU Group <gpugroup at gmail.com> wrote:
>
>
> https://www.web3d.org/documents/specifications/19774/V2.0/MotionDataAnimation/ExampleMocapAnimationMotionObject.html
>
>
> This version of AnnexD is what I used - and it looks like an SFString 0 as
> does values and jpints
>
> WARNING - any examples from this AnnexD may need rework for web3d
>
> -Doug
>
>
>
>
>
> On Sat, Jun 20, 2020 at 12:21 PM Don Brutzman <brutzman at nps.edu> wrote:
>
> MUFTI you definitely want to be looking at the HAnim2 specifications.
>
> Web3D Consortium members are the controlling authority for X3D and HAnim
> specifications, International Standards Organization (ISO) certifies.
>
> X3D3 is simply implementing HAnim2, typically as tersely as possible so
> that there is no confusion in prose that HAnim2 is the controlling
> specification.  Please see
>
> * ISO/IEC 19774-1, HAnim2 Specifications parts 1 and 2
>    https://www.web3d.org/documents/specifications/19774/V2.0/
>
> ===============================================
> * ISO/IEC 19774-1, HAnim2 Specifications part 1
>    clause 3 Terms and definitions
>
> https://www.web3d.org/documents/specifications/19774/V2.0/Architecture/glossary.html
>
> 3.11
>    level of articulation
>    LOA
>    degree of fidelity based on number of joints in an HAnim figure
> ===============================================
>
> also
>
> * 4.8.5 Levels of articulation
>
> https://www.web3d.org/documents/specifications/19774/V2.0/Architecture/concepts.html#LevelsOfArticulation
>
> "Level of articulation (LOA) represents the complexity and detail of
> joints for a humanoid skeletal hierarchy, and can be used for generating
> various motions based on the joints. There are five levels of articulation:
>
>      LOA‑0 represents only the humanoid_root Joint object without an
> accompanying hierarchy, as shown in (Figure 4.3).
>      LOA‑1 represents the simplest organization and hierarchy of joints
> for a humanoid. There are 18 joints and 18 segments. Each segment has a
> joint in the hierarchy. Figure 4.4 represents LOA‑1 joints.
>      LOA‑2 consists of 71 joints and 71 segments (Figure 4.5).
>      LOA‑3 consists of 94 joints and 94 segments (Figure 4.6).
>      LOA‑4 builds on LOA‑3 by adding anatomical details of hands and feet
> (Figure 4.7). LOA‑4 consists of 148 joints and 148 segments. Hands and feet
> joints for LOA‑4 are illustrated in 4.9.3 and 4.9.4 respectively."
>
>
> * 4.9 Structure of a humanoid, 4.9.6 Hierarchy
>
> https://www.web3d.org/documents/specifications/19774/V2.0/Architecture/concepts.html#Hierarchy
>
> * 4.9 Structure of a humanoid
>
> https://www.web3d.org/documents/specifications/19774/V2.0/Architecture/concepts.html
>
> * Annex A (informative) Nominal human body dimensions and levels of
> articulation (LOAs)
>    Nominal human body dimensions and levels of articulation (LOAs)
>
> * A.2 Levels of articulation (LOAs)
>
> https://www.web3d.org/documents/specifications/19774/V2.0/Architecture/BodyDimensionsAndLOAs.html#LevelsOfArticulation
>
> There is a lot of information in there that deserves careful reading.  You
> will find tables and diagrams that illustrate every joint and (bone)
> segment in the human body, along with a number of useful feature-point
> sites.  This took years of effort by many members of HAnim working group.
>
> Current work by Joe Williams, John Carlson and myself is improving
> past-legacy examples to match HAnim2 capabilities with X3D4.  We hope that
> testing and verification will help.
>
> * HumanoidAnimation X3D Examples Archive
>    https://www.web3d.org/x3d/content/examples/HumanoidAnimation
>
> * X3D Tooltips, HAnimHumanoid loa
>    https://www.web3d.org/x3d/content/X3dTooltips.html#HAnimHumanoid.loa
>
> HAnim designers hope that by identifying both HAnimHumanoid models and
> HAnimMotion animations with the LOA supported, it will be easier to
> mix/match models and animations.  Since each LOA is a strict subset of the
> next LOA, some compatibility is possible across human LOA levels.
>
> * 4.9.7 Site and Segment relationships
>
> https://www.web3d.org/documents/specifications/19774/V2.0/Architecture/concepts.html#SiteSegmentRelationships
>
> It is not impossible to think that someday this work might even rise to
> the level of rigor needed for 3D HAnim models becoming part of medical
> records.  HAnim and X3D4 have the potential to help... everyone.
>
> Thanks for your scrutiny and helpful questions.
>
>
> On 6/18/2020 11:02 PM, John Carlson wrote:
> >
> > To answer your question, many parts of the government like to create
> acronyms.
> >
> > John
>
> Sorry John, can't blame government for this one.  (Wondering, are parts of
> the government controlling your keyboard?!)
>
> > On Fri, Jun 19, 2020 at 12:38 AM J. Scheurich <mufti11 at web.de <mailto:
> mufti11 at web.de>> wrote:
> >
> >     hI,
> >
> >     From
> >
> >
> https://www.web3d.org/specifications/X3Dv4Draft/ISO-IEC19775-1v4-WD1/Part01/components/hanim.html#HAnimMotion
> >
> >     SFInt32 [in,out] loa -1 [-1,4]
> >
> >     What does "loa" mean ? Fieldnames in X3D are often english words,
> >     but the dictionary leo.org <http://leo.org> has no results for
> "loa" 8-(
> >
> >     so long
> >     MUFTI
> all the best, Don
> --
> Don Brutzman  Naval Postgraduate School, Code USW/Br
> brutzman at nps.edu
> Watkins 270,  MOVES Institute, Monterey CA 93943-5000 USA   +1.831.656.2149
> X3D graphics, virtual worlds, navy robotics
> http://faculty.nps.edu/brutzman
>
> _______________________________________________
> x3d-public mailing list
> x3d-public at web3d.org
> http://web3d.org/mailman/listinfo/x3d-public_web3d.org
>
>
>
>
>
>
>
>
>
> _______________________________________________
> x3d-public mailing list
> x3d-public at web3d.org
> http://web3d.org/mailman/listinfo/x3d-public_web3d.org
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://web3d.org/pipermail/x3d-public_web3d.org/attachments/20200624/46c9440c/attachment-0001.html>


More information about the x3d-public mailing list