Project deliverable Open Access

WhoLoDancE: Deliverable 2.6 - Motion capture sequences and skeleton avatar

Even Zohar, Oshri; Brekelmans, Jasper; Aarts, Jochem; Heuklom, Markus


JSON Export

{
  "files": [
    {
      "links": {
        "self": "https://zenodo.org/api/files/05f06c5d-b9ed-4e7f-a2d8-52d694ab1123/D2.6_Motion%20capture%20sequences%20and%20skeleton%20avatar.pdf"
      }, 
      "checksum": "md5:3c7a99858fb86e83e0617f5136dc73e2", 
      "bucket": "05f06c5d-b9ed-4e7f-a2d8-52d694ab1123", 
      "key": "D2.6_Motion capture sequences and skeleton avatar.pdf", 
      "type": "pdf", 
      "size": 1254199
    }
  ], 
  "owners": [
    29325
  ], 
  "doi": "10.5281/zenodo.1069018", 
  "stats": {
    "version_unique_downloads": 18.0, 
    "unique_views": 38.0, 
    "views": 39.0, 
    "downloads": 18.0, 
    "unique_downloads": 18.0, 
    "version_unique_views": 38.0, 
    "volume": 22575582.0, 
    "version_downloads": 18.0, 
    "version_views": 39.0, 
    "version_volume": 22575582.0
  }, 
  "links": {
    "doi": "https://doi.org/10.5281/zenodo.1069018", 
    "conceptdoi": "https://doi.org/10.5281/zenodo.1069015", 
    "bucket": "https://zenodo.org/api/files/05f06c5d-b9ed-4e7f-a2d8-52d694ab1123", 
    "conceptbadge": "https://zenodo.org/badge/doi/10.5281/zenodo.1069015.svg", 
    "html": "https://zenodo.org/record/1069018", 
    "latest_html": "https://zenodo.org/record/1069018", 
    "badge": "https://zenodo.org/badge/doi/10.5281/zenodo.1069018.svg", 
    "latest": "https://zenodo.org/api/records/1069018"
  }, 
  "conceptdoi": "10.5281/zenodo.1069015", 
  "created": "2017-11-30T14:51:20.030407+00:00", 
  "updated": "2019-04-09T13:20:03.418420+00:00", 
  "conceptrecid": "1069015", 
  "revision": 4, 
  "id": 1069018, 
  "metadata": {
    "access_right_category": "success", 
    "doi": "10.5281/zenodo.1069018", 
    "version": "1.0", 
    "license": {
      "id": "CC-BY-4.0"
    }, 
    "title": "WhoLoDancE: Deliverable 2.6 - Motion capture sequences and skeleton avatar", 
    "language": "eng", 
    "related_identifiers": [
      {
        "scheme": "doi", 
        "relation": "isVersionOf", 
        "identifier": "10.5281/zenodo.1069015"
      }
    ], 
    "relations": {
      "version": [
        {
          "count": 1, 
          "index": 0, 
          "parent": {
            "pid_type": "recid", 
            "pid_value": "1069015"
          }, 
          "is_last": true, 
          "last_child": {
            "pid_type": "recid", 
            "pid_value": "1069018"
          }
        }
      ]
    }, 
    "communities": [
      {
        "id": "wholodance_eu"
      }
    ], 
    "grants": [
      {
        "code": "688865", 
        "links": {
          "self": "https://zenodo.org/api/grants/10.13039/501100000780::688865"
        }, 
        "title": "Whole-Body Interaction Learning for Dance Education", 
        "acronym": "WhoLoDancE", 
        "program": "H2020", 
        "funder": {
          "doi": "10.13039/501100000780", 
          "acronyms": [
            "EC"
          ], 
          "name": "European Commission", 
          "links": {
            "self": "https://zenodo.org/api/funders/10.13039/501100000780"
          }
        }
      }
    ], 
    "keywords": [
      "motion capture sequence", 
      "skeleton avatar"
    ], 
    "publication_date": "2017-06-30", 
    "creators": [
      {
        "affiliation": "Motek", 
        "name": "Even Zohar, Oshri"
      }, 
      {
        "affiliation": "Motek", 
        "name": "Brekelmans, Jasper"
      }, 
      {
        "affiliation": "Motek", 
        "name": "Aarts, Jochem"
      }, 
      {
        "affiliation": "Motek", 
        "name": "Heuklom, Markus"
      }
    ], 
    "access_right": "open", 
    "resource_type": {
      "subtype": "deliverable", 
      "type": "publication", 
      "title": "Project deliverable"
    }, 
    "description": "<p>The&nbsp;conceptualization&nbsp;development&nbsp;of generic&nbsp;inverse&nbsp;kinematic&nbsp;(1)&nbsp;(IK)&nbsp;avatar skeletons&nbsp;for the&nbsp;WhoLoDance&nbsp;project&nbsp;was&nbsp;derived from the&nbsp;guidelines and requirements that emerged from discussions with the consortium dance partners&nbsp;from the project&rsquo;s start and through several dance partners and technical partners meetings in the last 14 month.&nbsp;&nbsp;</p>\n\n<p>The creation of a unified performer IK Skeleton Fitting (retargeting) and visualization&nbsp;was based&nbsp;on several principles manifested in those&nbsp;guidelines and&nbsp;pertained to the type of functionality an&nbsp;IK&nbsp;avatar&nbsp;skeleton&nbsp;should have in&nbsp;the&nbsp;project&rsquo;s&nbsp;context.&nbsp;&nbsp;</p>\n\n<p>For the avatar skeletons creation stage to be executed properly, there was first the need to&nbsp;analyse&nbsp;the motion capture sequences from the perspectives of:&nbsp;</p>\n\n<ul>\n\t<li>\n\t<p>Global scale deviations between&nbsp;performers&nbsp;</p>\n\t</li>\n\t<li>\n\t<p>Range of motions rotational scope per performer&nbsp;</p>\n\t</li>\n\t<li>\n\t<p>Unified mean deviation across all performers&nbsp;</p>\n\t</li>\n</ul>\n\n<p>Once this stage was complete, the data was used to determine the dimensions and internal scale of an empiric unified avatar skeleton that would fit the scaling of all the performers that were captured for the project.&nbsp;&nbsp;&nbsp;&nbsp;</p>\n\n<p>The stages of the skeleton creation pipeline involved 3D&nbsp;modelling&nbsp;of skeletal hierarchies, setting up correct biomechanical human limits rotations constraints for the skeleton and coding of transformation parameters so that motion capture data could be propagated correctly into the avatar skeletons.&nbsp;&nbsp;Furthermore, work was carried out in enabling the use of 3D geometry on top of the avatar skeletons,&nbsp;whether&nbsp;in a form of parent-child relation&nbsp;(2), or as a flexible envelope&nbsp;(3)&nbsp;</p>\n\n<p>This&nbsp;deliverable describes&nbsp;the&nbsp;processes of&nbsp;creation and implementation of inverse kinematic skeletons for the avatars&nbsp;modelled&nbsp;in T2.4, next to the creation of materials, textures, lighting setup and&nbsp;shaders&nbsp;for the models. This&nbsp;deliverable&nbsp;also includes preliminary&nbsp;results of&nbsp;real-time testing of those assets.&nbsp;</p>\n\n<p>In the upcoming WP6, the data will also undergo a stage of 3D modelling optimizations of avatar for volumetric / holographic projection and optimization for alternative projection methods and systems. The last part of this task involves the skeleton&nbsp;fitting&nbsp;(retargeting to fit the anatomy and morphology of the 3D avatars with the human performers that&nbsp;will be captured in the future), optimization of the inverse kinematic skeleton, and, finally, optimization of materials, textures, lighting setup and&nbsp;shaders&nbsp;for real-time interactive display.&nbsp;</p>\n\n<p>(1):&nbsp;Inverse kinematics\u202fis the&nbsp;Mathematical&nbsp;process of recovering the movements of an object in the world from some other data, such as a film of those movements, or a film of the world as seen by a camera which is itself making those movements. This is useful in robotics and in film animation. In robotics, inverse kinematics makes use of the\u202fkinematics\u202fequations to determine the joint parameters that provide the desired position for each of the robot&#39;s\u202fend-effectors.\u202fSpecification of the movement of a robot so that its end-effectors achieve the desired tasks is known as\u202fmotion planning. Inverse kinematics transforms the motion plan into joint\u202factuator\u202ftrajectories for the robot. Similar formulae determine the positions of the skeleton of an\u202fanimated character\u202fthat is to move in a particular way in a film, or of a vehicle such as a car or boat containing the camera which is shooting a scene of a film. Once a vehicle&#39;s motions are known, they can be used to determine the constantly-changing viewpoint for computer-generated imagery of objects in the landscape such as buildings, so that these objects change in\u202fperspective\u202fwhile not themselves appearing to move as the vehicle-borne camera goes past them. The movement of a\u202fkinematic chain, whether it is a robot or an animated character is&nbsp;modelled&nbsp;by the kinematics equations of the chain. These equations define the configuration of the chain in terms of its joint parameters.\u202fForward kinematics\u202fuses the joint parameters to compute the configuration of the chain, and inverse kinematics reverses this calculation to determine the joint parameters that achieve the desired configuration.</p>\n\n<p>&nbsp;(2):&nbsp;Parent-child relational&nbsp;modelling\u202fIn the world of 3D, users are able to organize their scenes by creating a hierarchy. The hierarchy is created through the process of parenting objects to one another from inside the program. When an object becomes a Child of another object (Parent), it will follow all transformations applied to the Parent. This is useful in the case where a character or 3D object has multiple parts and needs to move around in the scene. That way you only need to animate the Parent model and the Child objects will follow automatically.&nbsp;</p>\n\n<p>&nbsp;(3):&nbsp;Flexible envelope&nbsp;Avatars can be enveloped (skinned) to a skeletal rig. The skeletal rig is then animated with keyframes, or in the case of&nbsp;Wholodance, driven by motion capture data. This animation in turn, deforms the envelope.&nbsp;&nbsp;</p>"
  }
}
39
18
views
downloads
All versions This version
Views 3939
Downloads 1818
Data volume 22.6 MB22.6 MB
Unique views 3838
Unique downloads 1818

Share

Cite as