Native face recognition
Decrease risk and increase flexibility with the ROC SDK, a fully native face recognition solution from Rank One. Keep your data on your machine, no internet connection required.
Expand your software's capability by leveraging the ROC SDK. Face detection, search, verification, clustering, demographic estimation, and appearance classification are all available out of the box.
Any data, anywhere
Rank One algorithms excel under a multitude of facial variations. Whether faces are occluded, poorly lit, or have a unique expression, your data can be used.
Face tracking and video-based face recognition allow the ROC SDK to be easily deployed regardless of the data capture method.
Ingest data rapidly
Waiting days to process your data is a thing of the past. Rank One algorithms allow you to enroll approximately 10 images/frames, per CPU core, per second.
Matching speeds of approximately 20 million comparisons, per CPU core, per second, make realtime results the new standard.
Do more, with less
Templates generated using Rank One algorithms are an order of magnitude smaller than the closest commercial competitor1. What once required a server room can be done on a laptop or mobile device.
Industry leading accuracy
Compared against a widely used face recognition algorithm2, Rank One algorithms deliver significantly superior accuracy. Indeed, Rank One algorithms are among the most accurate on the market.
At a fraction of the cost of competitors, the ROC SDK makes high-quality face recognition available to all. At no extra cost, the ROC SDK is bundled with a demo application allowing rapid testing and evaluation.
Supporting your mission
Choosing the development language that best fits your needs is easy; the ROC SDK provides a C API along with Python, Java, and C# wrappers and is supported on most operating systems. 24/7 support by U.S. based engineers who built the technology allows your team to rapidly integrate and deploy the ROC SDK.