A recent study published in the Journal of Orthopaedic & Sports Physical Therapy (JOSPT) investigates the relationship between early sports specialization and hip kinematics in ice hockey goaltenders. Using Theia3D markerless motion capture technology, researchers examined how hip movements differ between early specialized (ES) and non-early specialized (NES) goaltenders. The findings suggest that ES goaltenders demonstrate distinct hip movement patterns, potentially as a strategy to mitigate injury risk. This research provides important insights into training practices, injury prevention, and sport specialization guidelines for young athletes.

A recent study published in PeerJ assessed the reliability of Theia3D’s markerless motion capture system under varying clothing conditions during walking. The findings indicate that both tight and loose-fitting clothing yield consistent gait analysis results, with minimal variation observed between trials and sessions.

It's a new era for Theia! See how our industry-leading AI-powered motion capture software is getting a fresh visual identity for 2025— reflecting our commitment to providing biomechanics professionals with the most advanced AI-powered motion capture tools trusted by researchers, clinicians, and sports scientists worldwide.

Check out this podcast on the next generation of biomechanics featuring Dr Scott Selbie

Check out this podcast on the past, present, and future of pose estimation featuring Dr Scott Selbie

The new, highly-anticipated Theia3D release we've been discussing here these past few months is now available for all Theia customers, and to commemorate this exciting release we're giving it a name; Theia3D Apollo! In this blog post, we’ll cover some of the new and exciting changes that come with Theia3D Apollo and provide the relevant links to previous blog posts.

This blog post outlines our approach to R&D and includes links to many peer-reviewed articles using or citing Theia3D.

In this blog post, we outline the newest TMBatch features introduced through the upcoming Theia3D Apollo software release.

In this blog post, we outline the newest features introduced through the upcoming Theia3D Apollo software release.

Theia3D uses a large number of precisely predicted keypoints and an inverse kinematic approach to obtain whole body pose estimates during any movement - but how much is the solution dependent on the IK model and its constraints? In this blog post, we investigate this question by comparing kinematics produced with varying degrees of freedom in the lower limb joints, and find that the kinematic results are largely unchanged - a nod to the robustness of our model and its keypoints.

The Theia3D Apollo update is coming soon! Theia3D kinematics have been evaluated against marker-based data many different times, including against varying model definitions. One of the most apparent differences that has come up during these evaluations are those in the pelvis and hip kinematics due to Theia3D’s use of a neutral pelvis. In this blog post, we continue our lead-up to Theia3D Apollo later this month by discussing the differences and changes that have been made to the Theia3D model which improve on the pelvis and hip kinematics, and go further to improve hip joint center positioning.

Theia3D's 2024 update, Apollo, is fast approaching! Please stay tuned over the next two weeks as we will be releasing multiple blog posts detailing some of the improvements to look forward to later this month! In this blog post, we provide an update on the validation of Theia3D Apollo against Theia 2023 and our reference marker dataset. This includes an examination of numerous biomechanical measures such as segment and joint angles, and shows subtle but meaningful improvements in several measures across multiple movement types, including slow, fast, cyclic, non-cyclic, bilateral, and unilateral movements. For interested readers, we recommend keeping an eye on the blog for more Theia3D Apollo update posts including a discussion of model changes and the effects of varying degrees of freedom on Theia markerless tracking. Stay tuned!

In this blog post, we discuss how to optimize your markerless motion capture system for multi-person data collections. We will also quickly review the preferences menu within Theia3D and answer frequently asked questions related to multi-person tracking situations.

In this follow-up post to our Characteristics of Video Data for New Users post, we’ll highlight some of the important lighting factors to understand so that you can get the best data from your collections.

In this blog post we'll discuss how CONTEMPLAS video systems work together with Theia3D to perform seamless markerless motion tracking.

As you may know, Theia3D utilizes synchronized color video data in order to track human movement. Naturally, it works best with high-quality video data! As a new user of markerless motion capture software, you may be unfamiliar with video data and its various characteristics, but in this blog post we’ll give you a quick overview of the aspects of video data that are important to consider when recording data to analyze in Theia3D.

New research on the repeatability of gait outcomes in patients with knee osteoarthritis using markerless motion tracking has been published!

In this blog, Hochschule Offenburg describes how they use Theia3D to perform markerless motion capture-based animations!

In this blog post, we discuss how to integrate synchronized signals from peripheral devices with your Theia3D kinematics.

In this post we provide some pointers for Theia users who are new to Visual3D, and how to approach post-processing in Visual3D.

In this blog post, we provide some guidance for working with Theia3D data in Visual3D.

In this blog post Rob Kanko summarizes the latest publication on Theia3D from the Queen’s University Human Mobility Research Lab. You can also check out the full publication in the Journal of Applied Biomechanics.

In this post we cover the topic of occlusions in markerless motion tracking.

A breakdown of some of the more common issues with collecting data and how to troubleshoot them.

Avoiding bias in algorithm development and training is a top priority. We consider data, architecture, training procedure as well as testing as critical components, and continually expand on how we test and retest this process to ensure it improves over time. Our goal is to provide a system that can accurately and reliably measure human movement, and these, among other processes, are essential in recognizing and eliminating factors that can negatively impact signal quality.

It’s important and needs to be thought through. Doing a literature review, equipment assessment, pilot study, and more testing is a great starting point to find a frequency that best fits your experiment.

All data is stored and analyzed locally - the user is responsible for the entire data life cycle. Using the C drive (or “boot drive”) for short term storage, an SSD or network drive for medium term storage, and a mass storage device for archiving will ensure that the data is properly managed for analysis and re-analysis, allow the IT protocols and security requirements for sensitive data of the site to be followed, and limit the likelihood of data loss.

Our premium onboarding program will get you ready to launch your first markerless motion capture data collection quickly and efficiently - on your own schedule.

In this blog post we interview support engineer Rob Kanko and get his insight on the portability of markerless motion capture!

Thinking of running your own validation study using Theia3D? Let us share our experience and advice for getting the most out of your efforts.

A new blog post that lists best practices for data collection and analysis with our markerless motion tracking system!

Patch 7 of Theia 2023 is now live with some modest but helpful features! These features are located in the settings pane and provide added clarity on how people are identified and tracked within the volume. Okay, let’s dive in!

Theia 2023.3 is now available in your downloads portal and ready to go. This version of the software is a small but important update to the 2023 release. It contains added features and model improvements that increase the overall accuracy of the system!

You asked, we answered. We expended serious effort adding valuable features that we believe make the Theia user experience that much better. Thank you for all the excellent recommendations.

This is likely the fastest and most impressive consumer grade piece of hardware we have tested. The build quality itself is great, and the results show in the analysis speed. With Theia 2023, this new hardware is supported.

This is the first of our Theia release blogs and the message is simple: "Things were good, and now they are better". We see subtle, but meaningful improvements in challenging poses (3D position and orientation of the multibody model) with large ranges of motion, and we see reduced variability for many signals. We are committed to documenting and “re-validating” each release as standard practice, with a goal to continually make each release document more comprehensive.

Every year new research articles are published using Theia3D. Take a look!

A review of Vicon's mocap system setup with Vue cameras along with FLIR's Blackfly S cameras.

Tips for collecting accurate output data with Theia3D.

More new and exciting research validating markerless motion capture using Theia3D.

Everything you need to know about budgeting for a markerless motion capture system.

Insight into how the data used to train our algorithms allows us to track all kinds of individuals and conditions.

New original research article from Frontiers in Sports and Active Living validating markerless motion capture using Theia3D during boxing.

NaturalPoint camera system review. Part 3 in our 4-part series on using different camera systems with Theia3D.

Qualisys camera review. Part 2 in our 4-part series on using different camera systems with Theia3D.

First installment in a 4-part blog series assessing camera systems to consider for Theia 3D markerless motion tracking. Part 1 - Sony

An excerpt from a new publication in Frontiers in Human Neuroscience validating the use of Theia3D for reliable gait analysis.

An excerpt from a new publication in Frontiers in Human Neuroscience validating the use of markerless motion capture as a digital biomarker.

A firsthand look at how Theia's markerless technology is helping experts at Sanford Health improve the return-to-play process.

A common challenge in human movement measurement is occlusions. Read about how Theia approaches this problem in markerless tracking.

Does there have to be a tradeoff between speed and accuracy in motion tracking? At Theia, we strive to constantly improve both.

What should you consider when vetting new technologies for your research program? Here's our list of considerations for going markerless.

In markerless motion capture, camera placement can significantly impact the accuracy of results. Here is some information on system setup.

Improve your standard of motion tracking by making the switch to markerless. You don't have to leave old data behind, you can reanalyze it!

Some thoughts on how to choose the best camera for your markerless motion capture projects.

How Theia uses supercomputers to accelerate our data pipeline, avoid training bottlenecks, and run more experiments.

Learn how to use reliable technology, rather than your eyes, to make better biomechanical assessments.

Discussing accuracy and precision in biomechanics, and understanding what your system can and can't measure.

Advice on how to check your calibration without having to redo it, and how to find a process that limits the need to recalibrate.

Explore some of the benefits of scaling up data collection with markerless motion capture.

Data management in labs is crazy! Take some time and think about how you will name your files and get buy-in from your lab, it will make a huge difference.

How Theia is dedicated to running a litany of tests and validations during each new release cycle.

We miss you and want to connect! Hopefully this becomes a place for cool stories and information regarding biomechanics, machine vision and artificial intelligence.

How we approached validation work, with an attached white paper. Peer reviewed publications linked as well.

Be practical and reasonable; this isn't magic. We need feature rich images that are in focus with decent lighting. Another consideration is the comfort of the person being measured. Generally, normal athletic clothing is a good choice.

Smoothing signals is really important to remove unwanted artifacts. Be sure to understand the effect of your filter, test it yourself, and make sure to consult literature.

Yes, as long as your camera system can synchronize video data with the treadmill force signals, the data can be merged post-collection. Theia3D supports camera systems that integrate with Bertec, AMTI, and TreadMetrix treadmills.

Typically, eight cameras are adequate, though it depends on the occlusions created by the treadmill.

While Theia3D tests all compatible cameras rigorously, not every commercial camera is certified due to data sync and image quality issues. Unfortunately, GoPro cameras have not been certified. Check our list of compatible cameras for verified options.

Absolutely. As long as your native camera system can sync input signals from these devices, you can merge them with Theia3D’s C3D file, ensuring comprehensive data analysis.

Theia3D supports multiple calibration paradigms. You can use traditional wand calibration results or our proprietary calibration board, which we provide to you at purchase.

Absolutely! As long as the native camera system can sync input signals from these devices, you can merge them with the Theia3D C3D file.

This depends on your requirements and end goals. Consider speed of movement, frame rate, camera resolution, and integration with external devices.

Several cloud solutions exist. Please reach out to our sales team to discuss your specific requirements.

Theia3D has batch processing functionality built in to streamline analysis of large video data sets. This application shows real-time progress updates from our proprietary video analysis technology.

Theia3D analyzes all video data locally so you don’t have to worry about patients or athletes losing their data to us or a cloud provider. After downloading the software from our web portal you won’t need internet connectivity again. You can analyze videos, export the data and share a report with patients all while respecting their privacy and existing data sharing agreements.

Theia supports an open-reporting infrastructure whereby users can leverage existing software packages such as Visual3D, R, Python, Matlab, Excel, Nexus and QTM for reporting purposes. If you have specific questions about what types of reports our partners provide, please feel free to reach out to us here.

Theia3D analyzes all video data locally so you don’t have to worry about athletes losing their data to a cloud provider. You can analyze videos, export the data and share a report with athletes all while respecting their privacy and existing data sharing agreements.

Theia can export human pose data in .fbx file format for animation and/or game development.

Report generation and biomechanical analysis functionality is not built into Theia3D. To generate reports, we recommend using Visual3D (which is included in our packaged offerings) as it can automatically detect the data exported by Theia3D and build the required models. However, we also export .c3d files that can be used in other software.

Currently, Theia3D supports a single model definition that is scaled and applied on an individual subject basis. The model joint constraints (degrees of freedom) can be modified as follows:

Theia3D automatically performs gap-filling and smoothing of the 3D poses for all people tracked within processed movement trials. This is achieved using the Generalized Cross-Validation Spline method, which is equivalent to a double Butterworth filter with some additional advantages.

Absolutely! Keep in mind that a faster movement such as a fastball requires a higher frame rate than a slower movement such as walking. Ensuring the video images are crisp and provide feature-rich views of the athlete is key. For more information on how frame rate affects image quality and Theia3D output, check out this blog post here.

Currently, Theia3D only tracks humans.

The processing time required for any given trial will depend on the length of the trial, the frame rate used, the number of cameras, the number of people being tracked, and your computer hardware. Using a top-of-the-line consumer-grade desktop computer, Theia3D can achieve analysis speeds of approximately 200 frames per second (e.g. 1 trial with 8 cameras recording at 85 Hz for 4 seconds can be processed in approximately 14 seconds).

No, Theia3D ships with a companion application that allows you to batch analyze a list of trials without supervision or manual intervention. Prior to batch processing you can organize your data, assign calibration files, select your analysis settings, and modify the batch analysis trial list using built-in tools before beginning the batch analysis for fast and easy data processing.

No, Theia3D is a post-processing markerless motion capture solution and cannot perform real-time person tracking.

Theia3D will analyze any movement so long as there is sufficient visibility of the patient or athlete. This means you can use a predetermined set of tasks from literature or pick and choose which tasks you would like to include in the movement assessment!

System setup can vary quite a bit depending on the operator experience and personnel available. We would generally recommend budgeting anywhere from 30-60 minutes for system setup and initial calibration. Tear down can take another 30 minutes which includes removing cameras from tripods and packing the system back into portable carrying cases. For more information on setting up a portable system, check out this blog post here.

Yes, you need software to run the cameras and collect the video data. This software is usually provided as part of a package with your camera system hardware. Theia3D processes the video data saved or exported by this software to produce 3D kinematic data (segments and rotation matrices) that is ready for analysis. Additional analysis or 3D processing software (such as Visual3D) can be used to interpret and report the 3D kinematic data.

The maximum sampling rate and capture length are limitations of the cameras used and thus are different for every camera system. Longer captures will take longer to analyze in Theia3D.

Theia3D can produce robust pose estimates for people in almost any environment, provided that the data quality and camera system setup meet our recommendations. Some environments where customers have recorded data for analysis in Theia3D include:

Yes, Theia3D can track multiple participants, provided that they are all sufficiently visible in the input video data. The number of people that can be tracked is dependent on their visibility, which is largely determined by the number of cameras in your camera system. Any participant that is sufficiently visible in three or more camera views will be tracked, provided they are not excessively occluded by objects or other people. Under crowded conditions, some people may not be tracked if they are excessively occluded by other people.

In general, participants should wear body-fitted clothing that provides rich visual features and allows the different parts of the body to be discerned from each other.

The capture volume size will be limited by the number of cameras. Larger volumes and volumes of more complex shape will require more cameras to provide adequate coverage.

Theia3D requires six or more synchronized video cameras to calibrate the camera system and produce pose estimates of the people visible in the videos. Video trials with fewer than six cameras cannot be loaded or analyzed using Theia software.

Theia3D is a markerless motion capture solution that outputs kinematic pose data in .c3d, .fbx or .json file formats using synchronized videos as input. If your camera system can record videos that are synchronized with external devices, then the kinematic outputs from Theia3D will be synchronized with the signals from your other external devices. If the signals from those devices can be exported to separate .c3d files, it is possible to merge the kinematic .c3d files from Theia3D with the .c3d files containing other measurements in software such as Visual3D.

Yes you can, however they must collect fully synchronized, high quality video data. We only support data collected using specific camera systems that we have verified the quality and synchronization of the captured videos. So, while you can use your own camera system, if it is not on our list of verified equipment it is your responsibility to ensure the system meets the requirements of Theia3D and we will not be able to provide support for any issues related to data collection or quality.

Theia3D utilizes local computational power to apply its deep learning algorithms on video data to obtain 3D pose estimates. Therefore, Theia3D has significant computational requirements which can be met with high performing consumer-grade computer components. Higher performing components generally reduce processing time and improve overall user experience.

Theia takes validation very seriously. We have an extensive repository of current validations performed using Theia3D which includes healthy walking, clinical gait, stroke, various neurological, ACLr and TJA patients. You can visit our blog or reach out to us here for more information.

Please reach out to our sales team to discuss enterprise deployments and bulk purchase discounts.

While we do not provide a direct API, some functionality exists to allow application communication between Theia3D and other systems.

While Theia3D is always branded as a Theia product, it can be run in the background of any commercial solution. In these cases, input from our engineering team is required to ensure seamless integration of technologies.

Because of our dedication to data quality for biomechanics, Theia3D is state of the art in terms of accuracy. To our knowledge, no markerless system has been validated to the extent of Theia3D.

Of course, we take validation very seriously. Validation consists of our own internal QA process and benchmark testing as well as third party validation at leading research centers across the world.

Theia3D is a markerless motion capture software that estimates 3D human pose from multiple synchronized 2D camera views using deep learning algorithms.

Theia3D does not require people to wear specific clothing to be tracked. However, loose or baggy clothing may result in lower quality tracking. As a general rule, if you can easily identify joints in the camera images, then Theia3D can infer the joint positions as well.

Ideal tracking conditions are achieved when all joints are visible in all cameras; however, this is rarely attainable due to occlusions from other limbs, people, and the environment.

Person identification and tracking perform best when the people of interest are fully visible and cover a large portion of the fields of view of the cameras.

The cameras must capture synchronous videos with identical start times and durations, or must be able to be timecode synchronized in order to be used with Theia3D.

Theia3D requires a minimum of six cameras for tracking; however, we recommend a minimum of eight cameras for most capture volumes.

Local segment coordinate systems can be toggled on and off using the sidebar button.

Skeletons can be toggled on and off using the sidebar button.

3D segments can be toggled on and off using the sidebar button.

Person identification boxes can be toggled on and off using the sidebar button.

Toggle showing the 3D View in the application. If toggled ON but the 3D View is not visible, drag the 3D View open from the right border of the application window.

Most of the Display options shown here can also be toggled ON and OFF using the Display toggle buttons in the sidebar.

The following video provides step-by-step instructions for setting up and using your cameras to collect data for use with Theia3D. This includes changing camera settings, synchronizing the cameras, recording data, and downloading videos from the cameras through the web interface. This tutorial demonstrates the basic functionality and recommended settings of the Sony camera control interface. Refer to the Sony documentation for information about other settings and more advanced functionality.

The following video provides step-by-step instructions for setting up your Sony RX0 II camera system for the first time. This includes configuring the cameras and control boxes, setting up the multi-camera system, and controlling the system through a web browser.

Each Sony RX0 II camera includes the following...

This section describes the basic camera system requirements for recording data to be used with Theia3D. See Data Collection for additional recommendations and principles to follow when recording video data for Theia3D.

The preferences file associated with each trial determines the prefences used when that trial is analyzed. This allows you to customize the preferences used across different trials within a single batch analysis, providing greater control over the analysis process. If there are not any preferences .pxt files assigned to the trials within your selected batch analysis root folder, the Preferences column will indicate that the default preferences will be used for your trials.

The Trials list is populated with all valid trials within the selected batch analysis root folder, once a folder has been selected. Above the trials list are the following options:

The Settings section provides widgets to set up the current batch analysis. The Browse button enables the user to select the root folder. The data to be processed must be organized in this single root folder that can contain as many levels of subdirectories as desired to organize the data. However, each branch of the directory must end with a folder containing the data for a single trial. This must be a folder of video data as described in Video Data that also contains the calibration files for the cameras. It is critical that the IDs of the cameras in the calibration file are the same as the IDs of the video file subfolders. The Settings section enables the user to input the batch analysis root directory and the output format that will be generated by the batch analysis:

Help‍ Opens the online documentation for the software.

The two prerequisites for batch processing of video data is that the videos are organized in the required nested folder structure as described in Video Data, and that every trial has an assigned calibration file. Therefore, the Organize Videos and Assign Calibration Files tools are included for use in TMBatch.

One of the benefits offered by Theia3D markerless motion capture is automated tracking, which allows it to analyze large datasets without human intervention or supervision. This is achieved using the TMBatch companion application to Theia3D, which allows a list of trials to be curated and batch analyzed sequentially. While batch processing is efficient and does not require supervision, we always recommend that you manually examine and check the quality of your markerless data and calibrations using Theia3D, before setting up a batch analysis. This can prevent poor calibrations or other issues with the data from going unnoticed until after the batch analysis has been completed.

Movement data for tracked people from a processed trial can also be exported to .json file format for more open-ended use in scripting environments and as an easily readable file output format. Use the Save Skeleton Poses button to save .json output files, or navigate to Save Skeleton Poses under the File Menu.

Pose files can also be saved as .fbx file formats for use in animation and other software tools that utilize this file format. Use the Save Skeleton Poses button to save skeleton pose .fbx files, or navigate to Save Skeleton Poses under the File Menu.

Pose (.c3d) files can be saved and used to perform post-processing analysis steps in Visual3D software. Use the Save Skeleton Poses button to save skeleton pose .c3d files, or navigate to Save Skeleton Poses under the File Menu. Saved pose .c3d files can be opened in Visual3D, and a subject-specific model will automatically be applied to the Theia3D data in Visual3D without requiring the model to be defined.

A workspace folder contains the video and data files of a saved workspace. The video files are named according to their unique camera ID. The .t3d and .p3d files contain the analysis results. It is important that the contents of the workspace folder are not modified, including moving files in or out of the folder.

The camera calibration file contains the calibrations for all of the cameras using a structure similar to XML. An example calibration file is provided with the sample data. Key elements and attributes of the calibration file are outlined here. Note that Qualisys calibrations exported from QTM as .txt files and Vicon calibrations (.xcp) can also be used.

The video data for a single trial must be contained in its own folder, and each video file must be in its own subfolder. The name of each subfolder must be the ID of the corresponding camera. There are no requirements on the structure of the names of the video files, but the names must be unique and videos must be .avi or .mp4 format. For example, the figure above shows video data for a walking trial collected using four cameras with IDs 21375, 21379, 21380, and 21381.

The full body kinematic model consists of one kinematic chain comprised of the lower body, upper body, and head, with the pelvis as the root segment. Abdomen and neck segments are included. The Full Body Model must be used in order to save skeleton poses as FBX format.

The Default kinematic model consists of two kinematic chains for the lower body (pelvis and legs) and upper body (torso, arms, and head).

Model pose can be exported to .c3d, .fbx, and .json files.

There are currently three available models in Theia3D

Help Opens the online documentation for the software.

The Setup preferences pane contains options and parameters for the software setup and startup.

The rendering preferences pane contains advanced options and parameters for how the data should be played back and rendered.

The Analysis preferences pane contains advanced options and parameters for adjusting how the movement is analyzed, including subject identification and person tracking, model parameters, and 3D reconstruction parameters.

Shortcut: Crtl+, The settings available at the top of the preferences pane enable the user to load, save, and save default preferences.

The Video Enhancement tool allows you to improve video quality before processing data, which can be especially useful for calibration under challenging lighting conditions. Changes made using this tool are previewed for the loaded 2D views, and any enhancements are applied before running the 2D inference, lens calibration, or chessboard calibration processes. Therefore, what you see in the 2D views is how the video data will be processed.

The Display Video Metadata tool can be used to easily review relevant metadata for the videos currently loaded, including:

The Toggle Views tool can be used to turn on and off individual camera views for the currently loaded trial. When turned off, camera views are greyed out in the 2D viewer area, and are not used when analyzing the trial.

The Modify People IDs tool allows you to swap the unique person ID numbers assigned to each person when multiple people are tracked in a trial. Click and drag an ID and drop it onto another ID to swap those two ID numbers. This change to the IDs is immediately visible in the 2D and 3D views. You can also right-click on an ID to display a context menu with options for removing that tracked person or making it person 0. Note: removing a person cannot be undone (except by re-running Track People).

The assign calibration file tool is used to copy and nest an extrinsic calibration file within individual trial folders, one level above the video files. This is best practice for storing calibration files with their associated movement trials, and is a requirement for running batch analyses.

The Format Sony Multicam tool is alternative approach to organizing video data recorded using a Sony RX0 II camera system, rather than using the Organize Videos tool. This tool can be used to quickly organize a dataset based on the video timecodes rather than the C000# trial IDs appended upon downloading, which is particularly useful for cases where the C000# trial IDs are misaligned between cameras for the same trial recording.

The Organize Videos tool converts a folder of videos into a structure that can be used by Theia3D. This functionality is useful when collecting large data sets.

Synchronization of the video images can be checked after the trial has been analyzed fully. For each view, the 2D representation of the skeleton in that view is compared to the 3D skeleton computed from all views. If the alignment for a given view can be improved by shifting its video sequence forward or backward in time, this frame offset is reported in the dialog. You can then choose to correct the synchronization by applying these offsets to the out-of-sync views and clearing the entire analysis.

The Adjust Calibration tool can be used to modify the position and orientation of the global coordinate system (GCS) after completing a chessboard or object calibration, or after loading an existing calibration file. The GCS projection on each 2D camera view and within the 3D View is updated live as the Position and Angle sliders are used to modify these parameters of the GCS localization.

Object calibration is performed in order to determine the position and orientation of all cameras in the system in 3D space using a recorded object calibration trials. See How-To: Record Extrinsic Object Calibrations for detailed instructions for recording object calibration trials.

Chessboard calibration is performed in order to determine the position and orientation of all cameras in the system in 3D space using a recorded chessboard calibration trials. See How-To: Record Extrinsic Chessboard Calibrations for detailed instructions for recording chessboard calibration trials.

Intrinsic lens calibration is performed using a calibration chessboard and is used to determine parameters associated with the camera lenses and to correct for distortion and other visual effects. Lens calibration is required for OptiTrack Prime Color camera users, and may be required for Qualisys and Vicon camera users who are not using those third parties’ wand calibration procedures. See How-To: Record Intrinsic Lens Calibrations for detailed instructions for recording lens calibration trials.

The extrinsic camera calibration can be checked after the trial has been analyzed fully. For each view, the 2D representation of the skeleton in that view is compared to the 3D skeleton computed from all views. If the alignment can be improved in at least 30% of the frames by translating the entire skeleton within the 2D image for that view, this information is reported in the dialog. You can then choose to deactivate the cameras that are out of calibration and clear the entire analysis.

The view command results window shows the commands that have completed by Theia3D while handling and processing data, including specific inputs and outputs in the dropdown menus when expanded. The text color indicates the step status, where green indicates success and yellow indicates a warning is present.

The batch progress dialog displays the progress of the currently running batch analysis. Each trial in the batch is shown and can be expanded to show the separate analysis steps. State icons (circles) are shown, and correspond to the same states as described in Trials.

Subject: The person to plot Euler angles for.Reference Segment & Segment: The Euler angles describe the orientation of the Segment relative to the Reference Segment, resolved in the Reference Segment coordinate system using a ZYX cardan sequence.Flip X/Y/Z: Flip the X, Y, or Z Euler angle plots.

Shortcut: Ctrl+F. Runs: Run 2D, Track People, and Solve Skeleton. Video data and calibration must be loaded.

Runs: Track People and Solve Skeleton. Run 2D must be completed.

Solve the 3D pose of each identified person. The kinematic model is scaled and the pose of the model is solved using inverse kinematics. A description of the model can be found in Default Model Description. Refer to Settings Menu for information about using a previously saved model to perform the inverse kinematics step. Track People must be complete before running Solve Skeleton.

Identify people across views (i.e., determine “who is who” in each of the views). Identification requries the person to be clearly visible in at least three views simultaneously. Each identified person is assigned a unique color that is applied to their boxes, segments, and skeleton. Run 2D must be completed before running Track People.

Detect people in each of the loaded videos. When complete, boxes are drawn around the detected people in the video overlays if Show Boxes is selected in the Display menu. Video data and calibration must be loaded before running Run 2D.

Cancel the currently running analysis. This option is available (and replaces all other options) only when analysis is running.

Save the 3D scene as an .avi video file. The video saved is identical to the 3D view shown in the 3D Viewer; therefore you should modify the options that affect the 3D visualizations in the Display Menu and Settings Menu before saving before saving 3D View videos. Select the desired frame rate and the frame range of the videos and press the Save button. Browse to the desired save location and enter the desired filename.

Shortcut: Ctrl+Shift+S, 2. Save each of the 2D view video overlays as an .avi file. The videos saved are identical to the overlays shown in Theia3D; therefore you should modify the options that affect the 2D visualizations in the Display Menu and Settings Menu before saving video overlays. Select the desired frame rate and the frame range (or select the option to use the analysis frames) of the videos and press the Save button. Browse to the desired save location and enter the desired filename. One video file will be saved for each overlay and the files are automatically named as filename_cameraID.avi.

Opens the pose data for all tracked people in Visual3D. Note: The filtered poses are loaded into Visual3D, and the path to the Visual3D executable must be set correctly in the settings dialog. We strongly recommend an active support agreement for Visual3D, as it is frequently updated to support the latest Theia model changes.

Shortcut: Ctrl+Shift+S, 1. Save the 3D pose of the tracked people. Select the person (or all people) and the frame range (or select the option to use the analysis frames) to save in the file. Files can be saved in .c3d, .fbx, or .json format. The data included in the files is described in Pose.

Save the scaled 3D model. To save the scaled model browse to the desired save location and enter the desired filename. Note that if multiple people are tracked a separate model will be saved for each of them. The files are automatically named as filename_personID.tmb, where personid is automatically assigned.

Save the current calibration. To save the calibration browse to the desired save location and enter the desired filename.

Shortcut: Ctrl+S. Save the videos, camera calibration, and analysis results in the current workspace to a directory in a format that can be opened by Theia3D. Select the frames to include in the workspace and press the Save button. Browse to the desired save location and enter the name of the directory that will be created to store the workspace files. To overwrite the current workspace, use the dialog to select a .t3d file in the existing workspace folder. Note that this only overwites the results (.t3d and .p3d files) and leaves the videos untouched. In this case, all frames in the workspace must be saved. It is not possible to save to an existing workspace folder other than the currently open one.

Open a previously saved workspace. To load the workspace, browse to and select the directory containing the workspace.

Shortcut: Ctrl+Shift+O – Load the camera calibrations for analysis. To load the calibrations, browse to and select the file containing the calibrations for all of the cameras. The calibration file must conform to the format described in Camera Calibration.

Shortcut: Ctrl+O. Load video files (.mp4 or .avi) for analysis. To load the videos, browse to and select the folder containing the videos. The structure of this folder must conform to the format described in Video Data.

Shortcut: Ctrl+F4. Clear all loaded data and analysis results.

Understand the interface design of Theia 3D.

Extrinsic object calibration trials are used to determine the position and orientation of every camera in your system relative to the desired global coordinate system. Extrinsic object calibration uses the static position of a calibration object with known dimensions or known positions of specific key points on the object. These 3D dimensions or positions must be measured with high precision. The calibration object should be sufficiently large or the calibration key points should be spaced far apart within the capture volume and most points should be visible in every camera view. The key points can be coplanar or can vary in all three global dimensions.

Extrinsic chessboard calibration trials are used to determine the position and orientation of every camera in your camera system relative to your desired global coordinate system. As with lens calibration trials, users with wand calibration-capable systems (e.g. Qualisys Miqus, Vicon Vue or FLIR Blackfly S systems) are not required to complete chessboard calibration. A minimum of one chessboard calibration trial should be collected every time you set up your camera system, but collecting multiple is recommended. A new calibration is required any time a camera is moved, so having multiple calibration trials collected throughout a long data collection session can help prevent data loss due to accidental or unnoticed camera movements.

Intrinsic lens calibration trials are used to determine parameters associated with the camera lenses and are used to correct for distortion and other visual effects. Lens calibration trials are required for all camera systems that make use of a chessboard calibration method, except for Sony RX0-ii cameras. Users with wand calibration-capable systems (e.g. Qualisys Miqus, Vicon Vue or FLIR Blackfly S systems) are not required to complete lens calibration unless they are planning to use the chessboard calibration method.

The following guidelines describe data collection best practices, but are not exhaustive. You may obtain high quality markerless motion capture data under different conditions from those described below.

Theia3D is a markerless motion capture solution that utilizes synchronized video data to produce accurate and reliable 3D pose estimates of humans visible within the video data. It leverages deep learning algorithms trained to identify humans and accurately predict the 2D positions of over 100 landmarks on the human body, in every video frame of every camera. By fitting a scaled subject-specific inverse kinematic model to the landmark predictions, the human’s pose is reconstructed in 3D and tracked throughout their movement. This data-driven approach results in a robust solution that is generalizable across environments and movements, allowing the accessible collection of high quality 3D motion capture data where it was previously impossible. Below we describe the basic framework for collecting markerless motion capture data using Theia3D; for more detailed instructions, please reference the appropriate section of this documentation and any accompanying videos.

The startup window appears if ‘Select GPUs on launch’ is enabled in the Theia3D preferences.

Theia installation steps. Login to the Theia Software Downloads portal using the login credentials provided to you via email.‍ Download the Theia3D application installer...

For any questions and issues, please reach out to support@theiamarkerless.ca

New users should start at the Getting Started guide. This guide contains tutorials that introduce you to the software interface and walk you through the most common workflows - such as calibration and analysis - so that you can harness the power of Theia3D for yourself.

Theia3D is the premier solution for accurate, reliable, and powerful markerless motion capture. Automatic tracking. Multiple people. Any environment. The only limitation is your imagination.

This error message arises when loading video data. It indicates that the videos files from the selected trial were written using an unsupported video chroma format for codec h264. Supported formats are 8, 10, or 12 bit YUV 4:2:0.

If Theia3D crashes when processing a calibration or movement trial, this issue is usually due to the GPU RAM becoming maxed out immediately when attempting to perform the processing. Typically, the root cause of this issue is too high of a GPU RAM requirement from the computer monitor setup, usually due to the use of multiple monitors or individual very high resolution monitors.

If Theia3D is failing to launch fully and is freezing after creating the main application window, this issue typically occurs after a change to the computer monitor setup.

If the calibration file is not visible when using the Load Calibration File button or the Assign Calibration Files tool and the calibration file is an .xcp file exported by Vicon Nexus, the file browser window may be filtering for .txt files only.

There are a few reasons why the skeleton may appear to be jittery when reviewing a processed movement trial

There are a few reasons why the skeleton may be completely missing when reviewing a processed movement trial

In general, the 3D skeleton appears to be incomplete when one independent part of the kinematic chain cannot be tracked. The body parts that are able to disappear will depend on the joint constraints you have selected in the Analysis section of the Preferences window, and the cause of their disappearance can vary.

There are several reasons why the projected 3D skeleton or body segments may appear to be incorrect when reviewing a processed movement trial

The most common reason for the projected 3D skeleton (or 3D body segments) to be consistently outside the body in the 2D videos is that there is an issue with the calibration file. The calibration of the camera system determines how the calculated 3D pose, represented by the 3D skeleton or 3D segments, is projected onto the 2D videos. Therefore, if there is an issue with the calibration file the 3D skeleton can be projected incorrectly onto the 2D videos, resulting in the skeleton or body segments appearing outside of the body.

If the global coordinate system is in an incorrect but consistent position and/or orientation in all camera views, this typically indicates that a different frame was used to set the origin than what was selected in the Chessboard Calibration dialog. This is often caused by the chessboard or its blue squares not being sufficiently visible in the selected Origin Frame, which can be a result of the chessboard being too far from the cameras, challenging lighting conditions, or the cameras being parallel with the surface of the chessboard. In this case, Theia3D searches for the nearest frame in which the chessboard is adequately detected for localization, and uses that frame instead, which can lead to a floating global coordinate system in an undesirable position and orientation.

If the global coordinate system is in different positions and/or orientations in each of the camera views, this indicates that an incorrect calibration file was loaded. If the loaded calibration file has camera IDs that match those for the loaded videos but the calibration corresponds to a different camera setup, the calibration file can still be loaded successfully. However, this will lead to the camera system being incorrectly calibrated and the global coordinate system will not appear in the location or orientation that is expected for the loaded video data. If this issue goes unnoticed and the trial is then processed, it will lead to the Track People Incomplete error dialog.

If the coordinate system is out of place in one camera view, but is positioned and oriented as expected in the remainder of the views, it is likely that the position of the single camera changed between the recording of the calibration trial and the loaded movement trial. The camera view may have been intentionally changed, as in the case of adjusting a camera view to better capture the volume, or it may have been accidentally changed, in the case of a tripod being bumped by a passerby.

This error message arises when abnormally high tracking errors are detected for one or more camera views, as reported in the error dialog. As indicated by the dialog, this usually indicates a problem with the camera calibration and may be possible to resolve using the Check Calibration tool.

This error arises when Theia3D is not able to adequately track the people identified within the 2D views throughout 3D space. There are two primary causes of this error:

This error message arises when the chessboard calibration trial does not contain sufficient frames of overlapping visibility of the chessboard in 3 or cameras throughout the trial. This prevents the camera system from being properly calibrated, as there is insufficient information to calculate the position and orientation of all cameras in the system in 3D space.

This error message arises when the chessboard calibration algorithm is unable to calculate the position and orientation of the cameras relative to the chessboard. More than one chessboard may have been detected in the calibration trial videos, which can be caused by extra chessboards present in the capture volume or mirrors positioned around the capture volume.

This error message arises when attempting to load a saved Theia3D workspace. It indicates that the trial folder selected when browsing to load a workspace is missing required files, such as video or .t3d files.

This error message may arise when loading a calibration file exported from Qualisys Track Manager after performing a wand calibration. It indicates that the selected calibration file has inconsistent field of view parameters for the cameras contained within the calibration file. The field of view parameters are important for determining the area of the camera sensor used in the recording of the calibration trial.

This error message arises when loading a calibration file. It indicates that the selected calibration file does not contain the required calibration parameters to calibrate the camera system of the loaded videos. The calibration file may be missing intrinsic parameters, extrinsic parameters, or a combination of both, which prevent it from calibrating the camera system for the video data.P

This error message arises when loading a calibration file. It indicates that the selected calibration contains camera IDs that do not match those for the loaded videos.

This error message arises when loading video data. It indicates that the videos files from the selected trial were written using an unsupported video file codec.

This error message arises when loading video data. It indicates that the videos from the selected trial are of unequal lengths.

This error message arises when loading video data. It indicates that the folder selected when browsing for video data to load contains fewer than six camera views.

This error message arises when loading video data. It indicates that the folder selected when browsing to load video data contains data that is not formatted as required by Theia3D (See Data Formats). The video data may all be located within a single folder (i.e. all video files together in one folder), instead of each video file being nested within its own folder with a matching camera ID folder name.

This error message arises when attempting to use the Organize Videos Tool to organize video files as required by Theia3D. Theia3D requires video data to be .mp4 or .avi file formats, so if the directory selected for organizing contains video files of a different format (e.g. .mkv, .mov, etc.) Theia3D will be unable to organize or load these videos.