FAQ

WE'RE HERE TO HELP

Get answers to common questions about the MoCap Pro Glove

Frequently Asked Questions

We've compiled a list of answers to common questions about our MoCap Pro gloves and how to get the most out of them. If you can't find the answer you're looking for please contact us directly for help.

General

The MoCap Pro Glove is a motion capture glove designed to capture fine hand and finger performance for use in animation across a range of industries including film and game development. It is a non-optical glove, instead using bespoke stretch sensors, which gives it a number of key advantages including the ability to record regardless of occlusion, dramatically reduced need for re-calibration, improved comfort and highly repeatable, accurate motion data.
StretchSense (part of Sensor Holdings) makes the MoCap Pro Glove using its bespoke stretchable sensors which are lightweight, robust, and provide highly repeatable data regardless of environmental fluctuations. For more about StretchSense, and to find out about its other product lines visit www.stretchsense.com.
Our soft stretchy sensors give unparalleled hand capture performance with no drift or need for re-calibration. We have also developed a pipeline for animators that puts art direction at the front of the process. These two combined shortens production time for studios and minimises costs.
Shipping times vary depending on location. Please contact us for the status of delivery times to your area.
Once your gloves have been delivered to you, please create an account on our website here. Next, our Support team will give you access to your license key, the Hand Engine download, and the registration link of our weekly Onboarding Webinars. You can check our Quick Start Guide to familiarize yourself with the software prior to the webinar session.
You can check out our Knowledge Base webpage, which is being constantly updated as we release new updates. If you can't find the answer you're looking for please please contact us at support@stretchsense.com.
Forward Kinematics (FK) is where you bend each joint in a finger by some angle and the tip of the finger ends up wherever it ends up based on the sum of all the angles in the finger joints. Inverse Kinematics (IK) is where you say "I want the finger tip to be here" —  at some position and orientation in space — and you work backwards from there to calculate what the joint angles should be and then set the joint angles based on this backwards calculation. Hand Engine's remapping tool does FK at the moment, we are working on an FK/IK solution which ultimately would mean if you are streaming data to a hand that is a different size and shape to the Hand Engine hand, having an IK/FK solution would enable a finger touch in Hand Engine to match a finger touch on the target hand. At the moment while it is all FK, if the hand is a different size/shape then a finger touch in Hand Engine may not produce a finger touch on the target hand and the fingers may be over or under rotated. (Updated December 2020 - contact us for the latest).

MoCap Pro Gloves

Our gloves are made of a stretchy material so have a good tolerance for various hand sizes and shapes. However, it is important that you purchase the right size glove for your actor or animator’s hands. We have sizes 1, 2, and 3. Generally speaking, most people will fit a size 2, while more women will find size 1 more suitable. Size 3 will suit men with large hands. See the MoCap Pro Glove Size Chart (PNG 371KB) for the exact dimensions of each glove size.  
Our gloves have a battery life of 8-10 hours continuously streaming data, enough  for a full day’s mocap shoot. Our modules require 2-3 hours to be fully recharged.
Yes, our gloves are made to be machine-washed. Pull the circuit and built-in battery out and the gloves are ready to be machine-washed and line-dried. This video shows how easy it is to sanitize the MoCap Pro.
Our gloves are designed for stunts including martial arts and use with props such as swords. Our sensors are very tough and because they are sewn into the glove fabric, therefore they cannot snap or break with reasonable use. Our sensors are currently being used in shoes, insoles, boxing gloves and medical devices for which robustness has been thoroughly tested. In addition, our gloves are machine washable and have surpassed 200 wash cycle testing in cold water with detergent. For further information about what constitutes "reasonable use" in the context of the MoCap Pro Glove please contact us
Each MoCap Pro Glove has five multi-segment splay sensors running down each finger and thumb. These sensors can measure the bend of the proximal interphalangeal joint and the metacarpophalangeal joints separately as well as lateral spread of the fingers (finger splay).
Our gloves can be calibrated using a process of mapping raw sensor values to various hand poses that you can choose.  Because our gloves fit snugly on your hands there is no a need to constantly re-calibrate the gloves. Watch a video demonstrating the stability of the MoCap Pro Glove.
The Bluetooth connection range for the glove is 30ft with ideal conditions (i.e. line of sight between glove and Bluetooth dongle). This can be affected/reduced by the positioning of the dongle, any obstructions in the way, including a performer's body or the body of another performer.
Ideally, dongles should be suspended from a powered USB hub on the end of an active USB cable from the middle of the ceiling of the motion capture stage. This will provide, in perfect conditions, a ~50-60ft Bluetooth network radius.

Hand Engine Software

Although Hand Engine opens in a web browser, you only need to be connected to the internet when you first activate your license and then at least once per 30 days after that to stay activated. Once activated, you can use Hand Engine offline just as you would any other software application. We can do offline activations on request, contact us for more information.
Hand Engine saves files according to the convention: <x>_<actor name>_<profile name>_Hand_<left/right>_<take name>_<date>_<time>.
You can live record information in MotionBuilder, Maya, Unity Unreal, and Xsens MVN. Live triggering record to SD Card is also available.
Hand Engine Animator is for sole operators and Hand Engine Studio suitable for studios. For a comprehensive list of features for each license type, see the Animator & Studio Comparison table on the Hand Engine page.
Captured data can be recorded in several locations to cover all use cases.  Raw data can be saved to a user-specified file on the SD card on board the gloves. Raw data, joint angles and your calibration profile are saved to a unique local folder on the computer every time you click record. Joint angles are also saved directly into the FBX of your character. Recording all of your parameters and data means that if you discover you are not happy with the calibration of your gloves the day after your studio shoot, you can update the calibration and reprocess all of the raw data again with the new calibration, thus avoiding a costly second session in the studio.
Hand Engine Animator supports 2 pairs of gloves at one time while Hand Engine Studio supports 6 pairs simultaneously.
We have developed a lightweight app that connects to a performer's on-body smartphone so you can broadcast data over WiFi to an off-stage computer. Ideal for a large mocap stage or multiple performers that are interacting with each other. You can also use our gloves in USB tethered mode (e.g. for desk animation). The app is currently only available for Android OS.
Hand Engine is capable of accepting software triggers but we have not established the documentation around this function yet. We intend to have pre-built support for common digital slate software, plus the ability to specify a generic input for custom solutions.
At the moment we rely on external 3rd party trackers for tracking hand location in VR applications, i.e. the Vive trackers. The finger angle data is streamed out of our Hand Engine software via TCP comms as a JSON formatted string that can be received and decoded by a range of software/devices with networking functionality.
There is a range of things that can affect the accuracy/precision of the gloves.
When using the gloves the user must first go through a calibration process, the quality of the calibration will in turn affect the final accuracy. Furthermore, the size of a user's hand/ fit of the glove will also add some difference in how the sensors are activated again giving variation to performance. Due to these and other contributing factors, we do not have a usable single number for accuracy.
This is a bit of a tricky one in some sense. To get the data from the sensors in the glove to the wireless Bluetooth dongle at the PC is typically in the range of 3.5 to 15ms. From there you will have additional latency internally in the machine before it is visualized on screen this can include things such as how the USB hardware is polled, how the programming of received data is buffered/managed, and even the screen refresh rate.

Third-Party Integrations

Yes, you can stream raw data from our gloves to build your own custom applications in Maya. See an example video of MoCap Pro gloves streaming into multiple platforms including Maya. Here's our guide about streaming to Maya.
Yes, you can stream raw data from our gloves to build your own custom applications in Unity. For example, watch this video of a virtual dominoes game using MoCap Pro Gloves, Hand Engine, and Unity. Here's our guide about streaming to Unity.    
Yes, you can stream raw data from our gloves to build your own custom applications in Unreal Engine. See an example video of MoCap Pro gloves streaming into multiple platforms including Unreal Engine. Here's our guide about streaming to Unreal Engine.
Software platforms such as MotionBuilder, Unity and Unreal Engine can combine streams of body and hand data on to a single character. Alternatively, you can record hand data separately and then merge it with body data in post-production.
See this video on how to integrate StretchSense MoCap Pro Gloves with OptiTrack body tracking in MotionBuilder.
You can timestamp the data by connecting an UltraSync One module to the glove. There is a range of options of feeding timecode into Hand Engine, such as Local System Time and Glove Clock and we are working on ways to accept timecode from additional sources. For more on configuring timecode in Hand Engine, see section 4 in the Hand Engine User Guide.
Yes, the Hand Engine version 1.1.0 does support Vicon timecode integration. Here's a quick guide on how to set it up.
In regards to time code syncing with any other motion capture systems, at the moment we are not integrated with any slates which are commonly used to triggering recordings from multiple sources.
We do have an HTTP post request that can trigger recordings remotely, but it takes some programming to make that work that we assist.
We would recommend streaming into another mocap system (e.g. Xsens, OptiTrack) and recording everything in there (hands and body combined in the one FBX). If you use an UltraSync One device you can synchronize the glove data recordings by timestamp after the fact.

How to

The MoCap Pro gloves have a Micro SD card slot in their electronic component for data recording. Here's a quick guide on how to setup SD Card recording in Hand Engine.
Hand Engine is always streaming time data with hand pose information to downstream software (e.g. MoBu). The "Key to Timecode" option in Motionbuilder should adjust the timeline to the time of day in the Hand Engine recording.
 
In MotionBuilder, there will be an option of inserting a keyframe at the point in time defined by the timecode stamp of that data packet (Key to TimeCode) or to insert a keyframe at whatever frame the MotionBuilder scene is on when the packet is received (Key to Timeline).
 
Please note a really common trap when using Key to TimeCode is that the hand animation will be inserted into the timeline at a point that is often very offset from the frame counter for the take itself. When you stop the recording and remind the take to frame zero and hit play, the hands won't do anything (because the FCurves/keyframes are offset in time). The critical step is to right-click on the Action timeline in MotionBuilder and select Time --> Frame take so that it moves the timeline to where the data is. As you may surmise this mode is really useful when you are combining multiple data sources that are synced to TimeCode, and if you are playing back a raw file recorded in Hand Engine, it will play it back with the original timecode data so it stays synced in time with your other data.
To jam-sync the glove hardware using an UltraSync One module you will need to plug it in via the USB-C port on the glove for a few seconds. You can jam sync the UltraSync One module to the house clock you have using a BNC/bayonet connector cable and you can use one UltraSync One module to jam sync all of your gloves.
Because glove sensors run up the back of the fingers and hand, make sure you only pull on the palm side of the glove. This video shows you the correct way to remove and put the gloves on.
You can remap your Hand Engine output to match the hands of a custom character. This will enable you to stream data to your custom character directly through Hand Engine. For detailed instructions on how to do this, see the Remapping with Hand Engine document.
You train them exactly the same way you train any actor. If the position of the knuckle is different from person to person there is significant room for overlap and zero impact on function as long as the knuckles fall somewhere within the sensing zone. The main issue with hand size is that you may need to move to a differently sized glove for actor comfort, should you switch talent.

Known Issues

Currently, the CSV files recorded via SD Card are being saved under an incorrect format where the order of columns is different, this causes issues when importing the data into Hand Engine.
We have put a short-term solution until a proper fix is in place:
1-  Download the below excel document that can do the column swap to correctly import the data.
2- Open excel document "Convert SDRecording.xlsx"
3- Open a recording from the SDcard (i.e. "GLOVE-1_Take015_2021-02-17_11-51-46-am.csv")
4- Copy the contents of this SD recording file into the Reference tab of the Convert excel document.
5- Move to the Output tab of the convert excel document.
6- Save the document as a new file with the .csv extension.
The newly created/saved file can be run through the Hand Engine file mode.
With recording to SD card, the glove just records the raw data from the sensors within the glove. This data can be passed through Hand Engine (HE) in order to reconstruct the hand motion by running HE in File mode. There are a few tricks to using the SD Card record function.
  1. The SD Card should be formatted as FAT32
  2. The SD Card cannot be hot swapped (i.e. the SD card should be inserted before you power on the glove and then only removed once the glove is powered down).
  3. When the SD Card is recording you should be able to see the light on the back on the glove alternate between blue and green (normally it just flashes blue).
Also there is no need to worry about the C (Charge) ->D (Data) switch, it just pertains to the USB power/communication settings.
You are plugged into the wrong port. Unplug from the USB-C port and plug instead to the Micro-USB port and make sure the side switch is set to C (charging) not D (data).
Can’t find what you need? Our customer care team is here.