Semcam Live performs AI vision processing on the camera, then handles real-time 3D reconstruction and skeleton output locally. With the HPE High-Precision Extension, it supports offline processing with keypoint error as low as 1.6mm. It can also work with Goku optical mocap in the same coordinate system, timeline, and project.
Semcam Live and ZVR motion capture systems are moving markerless capture from demos into professional systems that can be deployed, delivered, and operated over the long term.
Markerless motion capture is moving from content tools into life sciences, university labs, robotics training, sports rehabilitation, and large-space interaction. A usable system needs more than pose recognition: it needs stable synchronization, accurate calibration, low-latency output, reviewable data, rigid-body tracking, and deliverable interfaces.
Semcam Live uses synchronized multi-camera arrays to capture motion evidence from multiple viewpoints and reconstruct 3D motion in unified time and coordinates.
Each camera performs human detection, tracking, segmentation, 2D keypoints, ReID, and confidence estimation at the source. Active Center receives frame-aligned structured data.
Semcam Live can be mixed with Goku optical systems to simultaneously capture robots, props, tools, interactive controllers, training equipment, rackets, drones and other rigid objects.
For small studios, teaching demos, creative previs, and real-time capture of up to 4 people.
For life sciences, robotics, and animation/VFX, with support for the HPE High-Precision Extension.
A large-space flagship with 12-person capture and HPE support, built for multi-person data capture.
Every Semcam Live camera includes onboard AI compute for human detection, tracking, segmentation, 2D keypoints, ReID, and confidence estimation at the source.
The real-time pipeline transmits frame-aligned keypoints and confidence data instead of pushing multiple raw video streams to a central server, reducing network and host-computer load.
120fps real-time output supports on-site preview, interaction, and driving. The HPE High-Precision Extension supports offline processing after capture, with keypoint error as low as 1.6mm.
Semcam Live and Goku can be jointly calibrated with optical T-bars and L-frames, reaching calibration accuracy up to 0.1mm.
Semcam Live captures marker-free human bodies and Goku tracks robots, props, tools, interactive controllers, training tools and other rigid objects, both sharing the same project.
Real-time capture, 3D reconstruction, skeleton output, and HPE high-precision processing can all run locally, making the system suitable for projects with strict latency, data security, and on-site control requirements.
Subjects can be captured without markers or mocap suits. The real-time pipeline supports on-site preview and data checks, while HPE supports high-precision post-capture processing and can connect with OpenSim, Visual3D, C3D, mainstream force plates, EMG, and pressure-plate workflows.
Active Center brings real-time capture, HPE high-precision processing, Goku rigid-body tracking, and multi-protocol output into one workflow. The system supports interfaces and workflows including ROS, C++, Python, Matlab, MuJoCo, Isaac, OpenSim, Visual3D, C3D, Unreal Engine, Unity, and Blender.
Semcam Live is an edge computing AI markerless motion capture system. Target detection, 2D keypoints, ReID, tracking, segmentation, cropping and confidence calculation are completed on the camera side, and Active Center locally completes cross-camera matching, triangulation, 3D keypoints, skeleton solving, IK, filtering, retargeting and exporting.
Semcam Live is designed for multi-camera, simultaneous acquisition and on-site deployment, and does not rely on single video post-processing. The system uses a fixed-edge AI camera array to output frame-synchronized structured data and complete 3D reconstruction locally, making it more suitable for laboratories, robotics, virtual production, LBE and long-term training scenarios.
The live link delivers 120fps motion output and sub-100ms end-to-end latency, making it suitable for live previews, live drives, interactions and acquisition inspections. HPE is a post-harvest high-precision processing pipeline suitable for research, training data production, key shots and projects requiring higher quality keypoint results.
PRO is suitable for real-time capture of up to 4 people, and does not support HPE and facial capture; PRO+ is suitable for professional projects that require HPE for up to 4 people; ULTRA is suitable for larger spaces and multi-person capture, supports up to 12 people, supports HPE and facial capture, and will be launched soon.
Yes. Semcam Live is responsible for unmarked human motion, and Goku is responsible for rigid body objects such as robot joints, end effectors, tools, props, handles, training equipment, rackets, drones, and cameras. Both share the coordinate system, timeline, and project in Active Center.
It can be connected to OpenSim, Visual3D, C3D, mainstream force plate, electromyography and pressure platform processes, and can be extended with index calculation, action quality evaluation and data export through Semcam Biomechanics Plugin and Semcam Action Quality Assessment Plugin.
The system supports interfaces and workflows such as ROS, C++, Python, Matlab, MuJoCo, Isaac, Unreal Engine, Unity, Blender, MotionBuilder, Live Link, FBX, BVH, C3D, etc. The specific delivery will be configured according to the project scenario.
Yes. Cameras, Active Center, real-time reconstruction, HPE processing, project management and data export can all be completed locally, making it suitable for teams that have requirements for data security, experiment reproducibility, on-site stability and long-term deployment.
Human motion capture does not require motion capture suits, inertial sensors or reflective markers. If your project requires tracking rigid body objects, such as tools, props, robot ends, handles, or training equipment, you can use Goku for 6DoF rigid body tracking.
It is recommended to provide venue dimensions, number of people captured, type of action, whether HPE is required, whether Goku is required, target software or data format, real-time driver requirements, synchronization equipment and deployment cycle. The more complete the information, the more accurate the configuration recommendations will be.
Tell us your venue size, capture count, application scenario, whether you need Goku, target software, and SDK requirements. The Semcam team will recommend camera configuration, deployment approach, real-time/HPE workflow, and data interface options.
