4. MotionBuilder Plugin

Plugin Installation

  • Download the latest plugin version "NOKOV-MobuPlugin.XXX.exe, " which is compatible with MotionBuilder 2018-2022. Extract the plugin installer package, double-click the extracted plugin, and install it into the software directory. After the installation completes, click Finish to finalize the plugin setup. (The application of the glove human body in MotionBuilder is the same as below.)


MotionBuilder Setup and Usage

  1. Open MotionBuilder, click on "Resources—Asset Browser—Devices, " find "NOKOV-Optical Device, " and drag it into the Viewer window.

  2. Click the Online button in the interface at the bottom: "Navigator—I/O devices—I/O Seeker - Optical Device" to enter Live status. Click "Model binding, " choose the Create button, and under "Optical model, " click "Generate a new optical model." After playing, uncheck and recheck Live. Then, move the human body in XINGYING software, and the model in MotionBuilder will move in sync.

  3. Select "Define—Skeleton" in the right window, then click "Characterize Definition" in the Navigator. In the scene, select a joint, right-click to assign the skeleton in the definition to bind it. After all joints are bound, click the lock icon (Lock character) and select "biped" to complete characterizing.

  4. Import the model to be driven into MotionBuilder and perform the characterization process as described in step 4. In the "Character Controls" window, select the imported model character in the "Character" column, and select the XINGYING human skeleton character in the "Source" column. Allow the movement of the human body in XINGYING software to drive the model.


MotionBuilder Auto-Binding Feature

  1. When redirecting motion capture data to a skeleton, you can use MotionBuilder's auto-binding feature for one-click binding.

  2. First, connect the motion capture human body data. In Live mode, have the human body stand in a standard T-Pose and pause the motion capture playback. If using Edit Mode, pause the human body data at the first frame.

  3. Open the settings in the motion capture software, set the IP network card address to "10.1.1.198, " and select the "SDK" option. Connect the motion capture data in MotionBuilder by referencing the specific steps from the earlier "MotionBuilder Setup and Usage" section.

  4. In MotionBuilder's lower left-side "Navigator" tab, expand "I/O Devices," click on "I/O Seeker-Optical Device," and in the middle right "Information" section, click "Characterize" ;

  5. At this point, the motion capture human skeleton will be automatically bound in MotionBuilder. Expanding "Characters" will display the bound human body name, which will match the human body name in the motion capture data.


MotionBuilder Characterization Requirements

  1. When characterizing motion capture data in MotionBuilder, the imported model's skeleton should match the XINGYING human skeleton as closely as possible, and the hierarchy of the model's skeleton should align with the hierarchy of the XINGYING human skeleton. The hierarchy of the XINGYING human skeleton, skeleton names, and the skeleton prefix "Body3" refer to the XINGYING human body name.

  2. After importing the model into MotionBuilder, please adjust the model's pose to a T-Pose and ensure it matches the T-Pose of the XINGYING human body before driving the motion. The skeleton axes of the model must be consistent with the axes of the XINGYING human skeleton.


Creating Rigid Bodies in MotionBuilder

  1. Create a rigid body in XINGYING Live mode or load rigid body data in Edit Mode. Open MotionBuilder software, drag the XINGYING plugin into the scene, and click the Online button in the lower interface "Navigator—I/O devices—I/O Seeker - Optical Device" to enter Live status. Click "Model binding, " choose the Create button, and under "Optical model, " click "Generate a new optical model." Play XINGYING software, and marker points will be displayed in the MotionBuilder scene.

  2. In the lower-left "Navigator" tab of MotionBuilder, expand "I/O Devices, " click "I/O Seeker-Optical Device, " and in the middle-right "Information" section, click "Create RigidBody." The rigid body will then be successfully created.

  3. In the "Navigator" tab, expand "Scene, " and within "Scene" expand Seeker-Optical Device: Optical. Scroll to the bottom to see the created rigid body's name, which matches the rigid body name in XINGYING. Double-click the rigid body's name, and the rigid body in the scene will display connections. The selected rigid body will be highlighted, changing from red to green.


Display of Unnamed Points

  1. After obtaining motion capture data using the Motionbuilder plugin, if there are unnamed points in the motion capture data, these will also be displayed in the Motionbuilder scene. The default color for unnamed points is purple, while named points are colored blue.

  2. Adjusting the Color of Unnamed Points: In Motionbuilder, you can modify the color of unnamed points from the motion capture data in the scene. The steps are as follows: First, in the "Navigator" tab, expand "I/O Devices" and select "I/O NOKOV-Optical Device."

  3. Next, in the lower right corner of the Motionbuilder interface, select Properties. Below Default, click the "..." button, and in the popup Color window, you can modify the color of the unnamed points.


Scene Refresh

  • After receiving the human body data from XINGYING, if there is a change in the XINGYING template, you can select the "Information" tab at the bottom middle of Motionbuilder, and click the "Refresh" button to refresh the Motionbuilder scene. This prevents the human skeleton from becoming stuck in the scene when the XINGYING human body template changes.


Use Tpose

  • After characterizing the XINGYING human body data and the model that needs to be driven, it is necessary to select the created character and the source. Before doing so, we need to check the "Use Tpose" checkbox to force the XINGYING human body skeleton in the scene to display the standard T-pose posture. Since models imported into Motionbuilder are usually in a standard T-pose, we want our human skeleton to match the posture of the model. After ensuring consistency in the posture, then select the character and source. This way, when the XINGYING human body data drives the model in real-time, the model's actions will stay consistent with the XINGYING human body model's movements.