CapturyStudio is a software solution that uses computer vision and motion capture technology to analyze video images and extract data on human movement. It can be used with cameras directly connected to a computer or with previously recorded videos that are uploaded to the software.
CapturyStudio includes all the necessary tools, such as video stream synchronization, camera system calibration, actor model calibration, and tracking, to enable motion capture of humans with unprecedented flexibility. CapturyStudio has the capability to process an unlimited number of videos and track an unlimited number of performers simultaneously.
CapturyStudio is compatible with a range of video formats, including grayscale images, and can be used with various camera types such as GigE, HDMI, and SDI. Its ability to track grayscale images also enables the use of third-party IR cameras from Qualisys or Optitrack.
After capturing and processing videos, CapturyStudio enables users to stream 3D data to game engines or animation programs. Additionally, the software allows for data export in the most commonly used formats, including FBX, BVH, C3D, and CSV.
CapturyStudio offers comparable advantages to CapturyLive but at a more affordable entry price. It provides highly accurate and repeatable motion data that captures even the finest movements, including fingers. The software enables tracking of videos recorded in challenging physical and lighting conditions, all while being completely markerless.
Starting the tracking process with CapturyStudio is simple. Just place a 'skeleton' over a person visible in any of the video cameras and start the tracking process after a brief actor scaling. Repeat this process for additional performers. You can start the tracking at any point during the video sequence and track forward or backward. CapturyStudio can track as many people as are visible within the capture volume.
The Finger Tracking module is offered as an add-on for CapturyStudio. Finger Tracking is fully integrated with the full body tracking capabilities of the system and captures all finger segments.
CapturyStudio includes tools to perform temporal and spatial calibration for the cameras used during the recording process. This includes synchronizing cameras, the intrinsic and extrinsic calibration. In a typical project workflow, intrinsic and extrinsic calibration sequences are first recorded, followed by actor calibration and tracking sequences. The synchronization tool is used to perform temporal synchronization of multiple videos that have not been recorded with hardware-synchronized cameras.
To acquire and scale a person, a two-stage process is employed. Firstly, a skeleton is mapped onto the person in any video view. This is done by the operator dragging and dropping the skeleton on top of any person visible in the tracking area. In a second stage, the person is auto-scaled which helps to establish bone lengths and joint centers more accurately. The entire process only takes seconds.
CapturyStudy supports a variety of streaming protocols, with easy-to-use plugins available for Unreal, Unity, and MotionBuilder. Other supported streaming formats include BoB, OSC, VRPN, ROS, OSVR, and VIZRT. CapturyStudio also provides export options for FBX, BVH, C3D, and CSV formats, allowing you to export the data you need.
CapturyStudio provides a comprehensive selection of charting tools for motion analysis and visualization. It enables the measurement of 3D kinematics, including the rotation and position of all primary bones and joints, including fingers. The user has the flexibility to add, delete, or modify charts according to their needs. Force plates from AMTI or Bertec or EMG from Noraxon are supported.
CapturyDome is a 3D volumetric scanner that reconstructs models of the real world in motion from camera data. It produces realistic volumetric images of the captured subjects, highly realistic digital clones.
CapturyDome utilizes high-resolution 4k cameras to record humans, animals, or objects within a calibrated space from different viewpoints. Video data is streamed and saved to a cluster of camera servers for further processing.
The CapturyDome system captures both the geometry and texture of objects or subjects. It has the ability to reconstruct 3D models for each frame, providing a mesh sequence and video texture. It captures high spatial and temporal resolutions of dynamic scenes, enabling the accurate recording of movements and changes over time.
CaptureDome is used for research at the intersection of computer graphics, computer vision and machine learning. It enables the creation of algorithmic variations of what we see, essential for future intelligent systems, computer graphics, immersive telepresence, and autonomous systems.
CapturyDome empowers media companies to make the leap from 2D film to 3D VR. Live music venues can use live broadcasting into VR, allowing the audience to be on stage with the performer simultaneously. CapturyDome can be a valuable tool in the fashion industry, providing new ways to create, design, and fit clothing that are both cost-effective and highly personalized.
Control 100+ high-resolution cameras. CapturyDome's flexible distributed architecture allows capturing dozens of high-resolution high-speed cameras, both machine vision cameras and professional film cameras from a number of manufacturers including but not limited to XIMEA, FLIR, Matrix Vision, and RED.
Calibrating a system with many cameras works in the same way as with a smaller system. An interactive workflow guides the user through the calibration process. Typically, lens and volume calibration are performed separately. For Dome systems lens parameters are refined during volume calibration to give you the highest possible calibration quality.
Videos and calibration files can be exported from the distributed storage system and saved in a fully configurable folder structure in various formats incl. JPEG or PNG sequences, video files, Agisoft calibration files, etc. providing the flexibility you need to match the output to your pipeline.