This website uses cookies to improve your experience. OK
ROC FACIAL RECOGNITION
Is your system powered by the industry's fastest
and most accurate solution?
Learn More

The ROC SDK

A lightweight software development kit that gives system and device integrators access to the world's most versatile face recognition technology. Click on an icon below for more details.

ACCURATE


2x MORE ACCURATE ON WILD
7x MORE ACCURATE ON MUGSHOTS

FAST


6x FASTER ENROLLMENT SPEED
3x FASTER COMPARISON SPEED

COMPACT


SMALLEST TEMPLATE AND BINARY SIZE IN THE INDUSTRY

USABLE


BEST IN CLASS API WITH EASY SOFTWARE INTEGRATION

Industry leading accuracy

World leader in accurate facial recognition algorithms, as measured by the National Institute of Standards and Technology (NIST).

99.0%

Frontal
Constrained
(Photo ID)

98.1%

Frontal Unconstrained (Selfie)

92.5%

Non-Frontal Unconstrained (Wild)

Accuracy is measured as the 1:1 identity verification True Acceptance Rate at a fixed False Acceptance Rate of 0.01%. Source: NIST FRVT Ongoing. Amongst top-tier vendors, the ROC SDK is typically 2x more accurate on wild imagery and 7x more accurate on mugshots.

Unrivaled Speed

  • Fastest enrollment speed
  • Fastest comparison speed
  • Best option for video processing
  • Reduced server needs

170 milliseconds
Enrollment speed

Enrollment is the process of detecting faces in an image (or video frame) and creating "templates" that encode the identifying characteristics of each face. The faster the enrollment speed, the less computing power required to support a face recognition application. Amongst top-tier vendors, the ROC SDK is typically 6x faster.

340 nanoseconds
Comparison speed

The comparison speed is how long it takes to measure the similarity between two templates. The faster the comparison speed, the faster the verification or search results. Amongst top-tier vendors, the ROC SDK is typically 3x faster.

Speed is measured on Intel Xeon CPU E5-2630 v4 @ 2.20GHz using a single thread. Source: NIST FRVT Ongoing.

UNPARALLELED EFFICIENCY

  • Smallest template size
  • Smallest binary footprint
  • Best option for mobile environments


165 bytes
Template size

The template size is the amount of storage space required to save a template extracted in the enrollment process. The ROC SDK's small template size enables us to search large databases on a smartphone FASTER than some of our competitors can on a server.

155 MB
Memory footprint

Memory footprint is the total hard disk space that is required to store the ROC SDK software libraries. At 155MB, the standard ROC SDK memory footprint is one of the smallest on the market. At 7MB, the ROC Embedded memory footprint opens the door for mobile applications with powerful facial recognition capabilities embedded into the app and running on the edge device, without a data connection.

Easy to integrate

The ROC SDK is a lightweight, simply integrated software library that enables rapid system development.

Native multi-platform support

The ROC SDK is a fully native solution, with minimal system requirements and no internet is connection required. The ROC SDK supports all major operating systems and computer architectures. The ROC SDK exposes a C API with wrappers for Python, Java, C# and Go. A robust command line interface even allows constructing systems from shell scripts.

Android

macOS & iOS

Linux

Windows

ROC Technologies

Verification

Perform 1:1 verification with industry leading accuracy.

Identification

Search millions of faces on a mobile device, or billions on a server, in seconds.

Face Detection

Locate faces in photos and video with industry leading detection speed.

Landmark Detection

Track eye and chin locations.

Clustering

Group unlabeled faces by identity.

Face Quality Metric

Predict a face likelihood to generate a successful match.

Face Analytics

Analyze a face to predict its age, emotion, ethnicity and gender.

Video Support

Process videos in real-time using a single core on a mobile device or server.

Pose

Determine where a person is looking relative to the camera using roll, yaw and pitch measurements.

Liveness

Detect print and digital screen replay attacks using a single image.

Person Counting

Count persons in a video stream seamlessly.

Glasses

Detect eyeglasses, sunglasses, or no glasses.

Trusted by

6
U.S. DEPT. OF DEFENSE AGENCIES
20+
LAW ENFORCEMENT AGENCIES
5+
FORTUNE 500 FINANCIAL INSTITUTIONS
1 of 2
MAJOR GLOBAL CREDIT CARD COMPANIES
10M+
FACIAL VERIFICATIONS PER YEAR

Our latest blog posts

Learn more about facial recognition technology from experts in the field.

Contact Us

Please fill in the information below and one of our representatives will contact you as soon as possible.