Identity Verification


Synthetic data for computer vision to enable more capable and ethical AI.

For software developers that build, train, test, deploy, and optimize computer vision ML models for physical security, video monitoring, and access control systems, the opportunities that synthetic data creates are considerable: Lower development costs, faster release cycles, more performant models, and improved safety outcomes. Synthesis Humans can be used to train biometric access control systems, and Synthesis Scenarios is a tool that lets ML engineers create labeled 3D data scenarios with multiple people for building CV models in security applications.

Home Security

Home security systems now offer camera setups that can be trained to identify known individuals vs. strangers. Biometric-based systems can differentiate between family members, service personnel (e.g. delivery man), and potential threats. Synthesis Scenarios allows CV engineers to create a broad set of data to ensure robust model performance across camera placements, environments, people and actions.
home security
Facilities & Access Controls​

Facilities & Access Controls

Making security images and video files useful for ML model training requires accurate human annotation, a time-consuming, labor-intensive, expensive process. Software developers can use Synthesis Humans to develop biometric access control models. Additionally, Synthesis Scenarios can be used for threat detection ML model development involving complex scenes and multiple people.

Public Spaces

Computer vision systems are being trained to recognize different behaviors in public spaces for a broad range of applications, including activity classification, threat detection, and pedestrian traffic analytics. Synthesis Scenarios lets CV teams create labeled 3D data scenes populated with synthetic humans that have realistic facial, body and hand motion across a wide variety of body types, camera angles, backgrounds, and environments.
Public Spaces
Camera Angle and Pose

Camera Angle
and Pose

Each product and camera system is distinct, and our platform enables the generation of representative data across imaging modalities and lens types, and with any camera position.

Ready for a demo or want to learn more? Get in touch.

boy is wearing headphones

Ready for a demo? We're here to help.