Motion Capture & Motion Tracking

MoCap, short for motion capture, is a technique used to digitally record movement. In art, it's a powerful tool that allows creators to translate physical gestures into digital data that can be used across many digital art forms. Mocap can be recorded or used in realtime in contemporary art, performance, and installation work. 

Motion Tracking is used to follow the movement of a specific object, body part, or point — often in real time — but not necessarily capturing full-body motion or saving a performance.

In short:
MoCap is about capturing a performance.
Motion tracking is about responding to movement.

Which Mocap/Tracking do I choose for what?


Optical Inertial AI/Camera-based Kinect Vive Trackers Vive Ultimate
Capture & Tracking Capture & Tracking Tracking Tracking Tracking Tracking
Optical; multicam & markers Inertial, IMUs (gyros + accelerometers) Markerless, AI + RGB/depth cameras or webcam RGB + depth sensor Hybrid with external IR  Hybrid inside-out tracking
Film, dance, precise animation Performance capture Web/mobile art, interaction, low-budget, prototyping
Installations, skeleton-based interaction		 Room-scale performance, VR puppetry (objects) Vr Avatar Puppetry, Untethered mocap, portable installations



Various types of Motion Tracking, a comparison

Kinect & depth camera's 
Vive Ultimate 
AI-based Motion Capture

Depth-sensing / markerless camera-based mocap

 

How it works:

  • RGB camera, infrared depth sensor

  • Tracks body skeletons in 3D space without any wearables.

Strengths:

  • All-in-one: depth + skeleton tracking

  • Works out-of-the-box with good body tracking

  • Widely used in interactive installations and prototyping

Limitations:

  • Limited range and lighting conditions

  • Skeleton tracking is less robust than pro systems

  • Requires a (windows) PC and specific SDKs 

In art, Kinect is great for:

  • Interactive performances

  • Visuals that respond to body movement

  • Multi-user installations

See more info on 3d Depth camera's here

Inside-out inertial tracking with onboard cameras and IMUs
(think of it as a hybrid between inertial and AI/vision-based tracking)

 

How it works:

  • Unlike earlier Vive Trackers that rely on external Lighthouse base stations, the Ultimate Trackers use two onboard cameras and IMUs to track their position in space independently.

  • They perform inside-out tracking, meaning they see the environment rather than relying on it.

  • Designed to work with Vive XR systems, but are also being adopted for standalone tracking in XR, motion capture, and performance.

Strengths:

  • No need for external base stations (fully wireless)

  • Much more portable and scalable

  • Accurate enough for many art/performance uses

  • Easier multi-tracker setups

Limitations:

  • Still relatively new — fewer integrations than legacy trackers

  • Limited support in open-source or non-Vive environments (for now)

  • Needs line of sight and light for the onboard cameras to function optimally

In art, Vive Ultimate is great for:

  • Untethered performer tracking

  • Object tracking in environments where base stations are impractical

  • Mobile or temporary installations where quick setup is needed

 

How it works:

  • Uses a single camera (or a small number of cameras) and AI algorithms to detect and track body, face, and hand movement.

  • Examples include:

    • MediaPipe (Google): Real-time pose estimation in 2D or 3D

    • OpenPose : Widely used for body landmark detection

    • Move.ai: Advanced multi-camera AI mocap, often used with smartphones

    • DepthAI / OAK-D/ Zedi: Cameras with built-in AI processors that provide depth and pose data

Pros:

  • No suits or markers needed — just a (web)camera

  • Low cost, often free or open-source

  • Quick to set up, highly accessible for artists and educators

  • Can be embedded into web or mobile apps

  • Good for gesture-based interaction, web-based artworks, or low-budget capture

Cons:

  • Generally less accurate than optical or inertial systems

  • Often limited to 2D or rough 3D estimation

  • Struggles with occlusion, fast movement, or unusual poses

  • Limited support for fine detail (like fingers or subtle facial expressions)

 

Various types of MoCap, a comparison

MoCap, short for motion capture, is a technique used to digitally record movement. In art, it's a tool that allows creators to translate physical gestures into digital data that can be used to generate or manipulate digital work.

What is MoCap?

Motion capture often involves placing sensors or markers on a person’s body (or using camera-based systems) to track movement in 3D space. This data is then sent to software that interprets the motion and applies it to a digital avatar, 3D model, or visual system.

Examples of use:

Why artists use it

There are various types of Mocap:

Optical Motion Capture
Inertial Motion Capture

How it works:

  • Uses cameras (usually infrared) to track reflective markers or colored dots placed on the performer.

  • Multiple cameras triangulate the position of each marker in 3D space.

 

Variants:

  • Passive optical (uses reflective markers + infrared light, e.g., Vicon or OptiTrack)

  • Active optical (uses LED markers that emit their own light)

 

Pros:

  • Very accurate spatial tracking

  • Excellent for large-scale and high-precision capture (e.g., dance, film, games)

  • Good for multiple actors and full-body motion

 

 

Cons:

  • Requires a studio setup with multiple calibrated cameras

  • Sensitive to occlusion (when a marker is hidden from view)

  • Expensive 

 

How it works:

  • Uses IMUs (Inertial Measurement Units), which are small sensors containing gyroscopes and accelerometers.

  • Sensors are worn in a suit (e.g., Rokoko, Xsens) and measure rotation and acceleration to calculate joint angles and movement.

 

Variants:

  • Can be combined with Optical Mocap for precision. 

 

 

Pros:

  • Portable: Can be used anywhere, indoors or outdoors

  • Not affected by lighting or line-of-sight

  • Great for live performance, field work, and small studios

 

Cons:

  • Less accurate in tracking absolute position (especially in large spaces)

  • Susceptible to drift over time (though software can correct this)

  • Locomotion is harder to grasp, like jumping, climbing etc.
  • Rokoko: frustrating glitches & subscription needed for realtime.


Some systems combine optical + inertial tracking (e.g., combining Xsens suit with camera tracking or facial capture or Rokoko, iphone & Coil ), giving the best of both worlds—especially for virtual production and advanced installations.

Using Motive and GazeboOSC for realtime OSC messages

The following tutorial explains the use of Motive with Gazebo for sending real-time Mocap data to other applications.

1. Preparing Motive 

To prepare Motive for sending internal NatNet data to Gazebo, go to the 'data streaming' pane in Motive and set the streaming destination to "Loopback" for streaming data wireless through 'Streaming Vlan' (ask your nearest blackbox manager for more info) or select a network switch for wired connection (The switch in the blackbox workshop at location Oudenoord is by default set to 192.168.10.30). The NatNet data can now be received in GazeboOSC (see pictures below)

Screenshot (11).png

Screenshot (13).png

2. Setting-up Gazebosc  

In Gazebosc you need to build a patch to convert Natnet data to OSC (see picture). Use the following actors by right clicking in the Gazebo workspace; 

Gazebo_1_green_chords.png

Once the correct connection is established between Motive and Gazebo the patch cords connecting the actors should colour green, an indication that data is streaming through Gazebo. To monitor the OSC data from GazeboOSC, or other OSC, data you can download the free OSC and MIDI monitoring application Protokol.  

The following example shows GazeboOSC distributing OSC data from ZigSim to different destinations.

A further explanation about Gazebo and it's use can be found at: https://bookstack.hku.nl/books/gazebosc

Rokoko

What is the Rokoko Suit?

The Rokoko SmartSuit Pro is a wireless motion capture suit that tracks full-body movement in real time. It's made up of sensors placed around the body, allowing you to capture the motion of a performer and translate it into digital animations.

What Can You Use It For?

The suit is designed for both recording and live-streaming motion data. This makes it ideal for:

It connects to the Rokoko Studio software, where you can see the motion data live, record takes, and export it to tools like Blender, Unity, Unreal Engine, or TouchDesigner.

Rokoko

How to Rokoko

important: NEVER firmware-update any part of  the suit without prior contact with the Blackbox!

image.png

The Rokoko SmartSuit Pro is a wireless motion capture suit that tracks full-body movement in real time. It's made up of sensors placed around the body, allowing you to capture the motion of a performer and translate it into digital animations..

What do you need to use it?

How to setup (steps)

Screenshot 2025-05-20 at 14.40.13.png

  1. Open Rokoko studio
  2. Create a Rokoko ID (in the studio, but directs youto the browser)
  3. Create a Project & Scene
  4. Create an Avatar, with your sizes (rough estimates can work measuring is better)
  5. Connect the Smartsuit to the computer. Use the provided USB-C cable & connect to the sensor on the back of the suit.

  6. Screenshot 2025-05-20 at 14.57.30.png

    Select the second icon to connect to your device (smartsuit)
    (If the suit does not appear, check your firewalls)
     
  7. Setup the wifi (preferably for the dedicated Router Specs**. )
    Use the 5ghz option if available.
  8. Connect the powerbank & disconnect the Usb-pc cable
  9.  Connect the actor profile to the device
  10. Wear the smartsuit (this step can be done earlier if you are working together)
  11. When using the gloves follow steps 5-9 again for each glove. Although the powerbanks we use has 3 outputs, it is preferred to use separate powerbanks for the gloves.

How to record mocap (steps)

  1. Callibrate
  2. Record
  3. Clean data
  4. Export 

How to live streaming 

For real-time data streaming you need a license/paid seat.
Info & prices: https://www.rokoko.com/pricing (set it to per month instead of anual> 28,- euro per month)
Loophole: Try it out for free: https://support.rokoko.com/hc/en-us/articles/4410424273169-How-can-I-access-the-free-7-day-trial-of-Studio-Plus-or-Pro 
For this you need to set up a team (https://support.rokoko.com/hc/en-us/articles/4410409137297-Creating-a-Rokoko-Team-and-Selecting-a-Subscription-Plan)

** The HKU Rokoko's MacAdress has been added to streaming Vlan so the suit also works in all of HKU & connects to your computer if you place it in streaming Vlan too. Connect with your local Blackbox Employee for help with this ;)

HubLights:

The sensors light up blue/green before turning off. When your Smartsuit Pro II sensors are in a normal state they will not be lit when powered on. The only lights that will be on during use will be that of the HUB.

image.png

Power LED
 What does it mean?
RED There is a problem communicating with some of the sensors(possibly a broken wire or sensor)
YELLOW Failed Redpine initialization
GREEN The power is on! The Smartsuit Pro should be detected in Rokoko Studio in the Device Manager
OFF
 Smartsuit is not connected to a battery

WIFI LED  What does it mean?
RED

A failure has occurred while trying to connect to the network. Please double check your WiFi settings(network/password/IP/Firewall etc) and reach out to support@rokoko.com if further assistance is required.

This colour is normal if you have changed computer or network or if this is the first time connecting your Smartsuit Pro to your network
BLINKING 
GREEN		 The Smartsuit Pro's Wifi function initializes. The Smartsuit Pro is also searching for WiFi

GREEN

The system is working properly and connected to an access point in the 5GHz band

BLUE

The system is working properly and connected to an access point in the 2.4GHz band

YELLOW

The system is working properly and connected to an access point in the Dual band mode

BLINKING PURPLE

The Smartsuit Pro's Hotspot is being initialized

PURPLE

The device is being connected to the PC via the Hub Hotspot
  

Issues & Troubleshooting

useful links:

Vive Tracking

Vive Trackers are small motion-tracking devices designed to bring real-world objects into virtual environments. While they are commonly used in VR setups to track body movements or props, they can also function as standalone tracking devices in software like TouchDesigner. Each tracker sends precise position and rotation data, making them useful for interactive installations, motion capture, and experimental media projects. By integrating them into creative software, we can map real-world movement onto digital visuals, audio, or other interactive elements.

Vive Tracking

Vive Trackers

image.png

What Are Vive Trackers?
Vive Trackers are small motion-tracking devices designed to bring real-world objects into virtual environments. While they are commonly used in VR setups to track body movements or props, they can also function as standalone tracking devices in software like TouchDesigner. Each tracker sends precise position and rotation data, making them useful for interactive installations, motion capture, and experimental media projects. By integrating them into creative software, we can map real-world movement onto digital visuals, audio, or other interactive elements.

Setting Up Vive Trackers with a Vive Pro & SteamVR

To use Vive Trackers standalone (without VR controllers), follow these steps:

1. Hardware Setup

A. Prepare Your Equipment

B. Position the Base Stations

2. Install & Launch SteamVR

3. Pair the Vive Trackers

  1. Turn on each tracker by pressing the power button until the light turns blue.

  2. Open SteamVR and go to Devices > Pair Controller (works for Trackers too).

  3. Select "Vive Tracker" and follow the instructions.

  4. If using a USB dongle, plug it into your PC and ensure the tracker is paired to it.  Place the extension cradles at 45 cm from the computer for stable tracking.

  5. Once paired, the tracker’s LED will turn green.

4. Configure the Trackers in SteamVR

5. Use in TouchDesigner 

6. Use in other software

Troubleshooting

Vive Tracking

Vive Ultimate Trackers

image.png

What Are Vive Ultimate Trackers?

The Vive Ultimate Tracker is HTC’s next-generation tracking device designed to improve motion tracking in VR and standalone applications. Unlike the regular Vive Tracker, which relies on external base stations, the Ultimate Tracker features inside-out tracking, meaning it uses built-in cameras to track its position without needing external sensors.

This makes the Ultimate Tracker more portable and flexible, as it doesn’t require a dedicated tracking area or base stations.  

Comparison: Ultimate Tracker vs. Regular Vive Tracker

Feature Vive Ultimate Tracker Vive Tracker
Tracking Method Inside-out (built-in cameras) Outside-in (Base Stations), infrared.
Base Stations Needed ❌ No ✅ Yes (Lighthouse)
Connection Type Wi-Fi 6E / USB-C Dongle USB dongle (2.4 GHz)
Battery Life ~7 hours ~7 hours
Size & Weight Slightly smaller & standalone Various sizes depending on the. model (1,2,3) & requires  base station setup
Latency Slightly higher due to processing Lower latency with base stations
Connection Built-In camera's, do not cover. Sometimes loses connection. Calibrates through mapping setup at startup. When connection is lost, move back to the centre point of callibration Base stations, do not cover the tracker for line of sight connection, then stable. 
Zero point connected to headset initial calibration  placement. This can be updated it TD
Best Use Case Portable, flexible tracking without a defined play area High-precision tracking in controlled spaces

When to Use Each?

  • Use the Vive Ultimate Tracker if:

    • You need standalone, flexible tracking without base stations & headset.

    • You’re working in large or dynamic spaces where base stations aren’t practical.

    • In a well-lit enviroment (but not too much direct sunlight)

  • Use the Vive Tracker if:

    • You need high-precision tracking with minimal latency.

    • You’re working in a controlled space with base stations.

    • You need compatibility with existing VR setups (e.g., Vive Pro).

    • When working in low-lit spaces.

1. Hardware Setup

A. What You Need

  • Vive Ultimate Tracker(s), preferably with straps

  • Usb -c Hub & extension cradle

  • PC (windows only) with Vive software installed

  • USB-C Cable (for wired mode, firmware updates & charging)

B. Charging & Powering On

  • Charge the Ultimate Tracker via USB-C before first use.

  • Press the power button until the LED lights up Blue.

2. Connect to Vive Wireless Hub

  1. Install & Open the Vive Wireless Hub Software

    • Download & install Vive Hub from HTC’s official site.

    • Launch Vive Hub on your Windows PC.

  2. Pair the Ultimate Tracker (link for steps on the Vive website, link)

    • Connect the USBDongle to your computer at least 45 cm away for decent tracking.

    • Open Vive Hub and go to the settings, choose Vive Ultimate Trackers

    • Check for firmware update & do them if necessary (app explains the steps) .

    • Pair New: Follow the on screen instructions to pair each Ultimate Tracker. 

  3. Tracker Mapping

    • click on Tracker setup & follow the steps on screen>
       

3. Open SteamVR

4. Use in TouchDesigner

5. Use in other software

Troubleshooting

Optitrack Motion capture

 

General Info

Example projects

Motion capture & Unity

The OptiTrack Unity3D Plugin allows you to stream real-time Rigid Body, Skeleton, and HMD tracking data from Motive into Unity. Using the streamed data, objects and characters in the scene can be animated.

All the information and plugins you need can be found here: https://docs.optitrack.com/plugins/optitrack-unity-plugin

Mind also these settings which are not so clear through the website stated above:

Settings for Motive & Unity

Connect both pc's with HKU netwerk Wifi or ethernet
Put mocap &  Unity pc in SVlan: ask blackbox employe to help, Restart connectios to make sure.
Ip range should be 10.3.4.xx (not 10.3.40.xx)  (3rd number is location based, 4 = IBB)

Disconnect all firewalls on both pc's! (Mocap pc is already disconnected)

Motive > view > Datastreaming Pane > Local interface: ip adres from wifi or ethernet verbinding
Transmission type: Unicast

Open Unity file
client optitrack (asset/game obj) connection settings:
Server adress: adres from mocap pc
local address: Unity pc

In Unity adjust Rigid Body ID, > Motive:
Assets > properties > Streaming ID

Camera (&AI) based Mocap

AI- and webcam-based motion capture is a form of markerless motion capture: instead of using suits with sensors or reflective markers, the system analyzes video images from one or more cameras and estimates the position of the human body using computer vision and machine learning. It can often run using ordinary webcams, smartphones, or consumer cameras.

How AI-Based Mocap Works

1. Video Capture

A webcam, smartphone, DSLR, or multiple cameras record a performer moving through space.
Depending on the system:

2. Pose Estimation

AI models analyze each video frame and detect key body points such as:

These points are often called landmarks or keypoints. The AI has been trained on massive datasets of human movement, allowing it to recognize body posture even under imperfect lighting or partial occlusion.

3. Skeleton Reconstruction

The detected points are connected into a digital skeleton or “rig.”
The software estimates:

Advanced systems may also track:

4. Retargeting

The motion data is transferred (“retargeted”) onto:

This allows live movement to control digital media from recordings or in real time.

Example Tools:

MediaPipe

Open Pose

FreeMoCap

MediaPipe is an open-source framework developed by Google for real-time AI perception.

It includes models for:

  • body tracking
  • hand tracking
  • face tracking
  • gesture recognition

MediaPipe is widely used because it:

  • works in browsers, Python, mobile apps, and game engines
  • runs efficiently on consumer hardware
  • supports real-time interaction
  • is relatively easy to integrate into creative coding environments

MediaPipe is often connected to:

  • TouchDesigner
  • Unity
  • Unreal Engine
  • Resolume
  • Blender

OpenPose

One of the foundational open-source AI pose estimation systems developed by Carnegie Mellon University.

Widely used in:

  • research
  • interactive installations
  • experimental audiovisual systems
  • dance technology
  • real-time

It can track:

  • full body
  • hands
  • fingers
  • facial landmarks

FreeMoCapis an open-source motion capture system focused on accessible, research-based full-body capture.

Unlike simple webcam pose estimation,
FreeMoCap can use:

  • multiple webcams
  • synchronized cameras
  • AI pose estimation
  • biomechanical reconstruction

It combines tools such as:

  • MediaPipe
  • OpenCV
  • scientific motion analysis workflows

The system triangulates body positions from multiple camera angles to reconstruct motion in 3D space more accurately than a single webcam setup.

mostly for recordings, not realtime

Move.ai

DeepMotion

Rokoko Vision

Move.ai

Uses AI and multiple cameras or phones for high-quality markerless mocap without suits. Often used in:

  • virtual production
  • indie filmmaking
  • game animation
  • previs workflows

Mostly processed capture, some live workflows emerging

https://www.deepmotion.com/ 

Allows users to upload ordinary video footage and generate motion capture animation automatically using AI.

Useful for:

  • rapid prototyping
  • avatar animation
  • virtual influencers
  • metaverse applications

Free version tracks up to 2 people, not realtime

Rokoko Vision

Originally known for inertial mocap suits, but now also supports AI webcam-based tracking.

Popular in:

  • indie animation
  • live performance
  • virtual characters
  • VTubing
  • interactive installations

Free for up to 15 seconds