FaceMe® eKYC & Fintech

FaceMe^® Accuracy

FaceMe^® Model

Ultra High Precision
(UH)

Very High Precision
(VH)

High Precision Model
(H)

Recommended Platform

Server, Workstation

PC, Premium Mobile

Mobile, Light-weight IoT/AIoT Devices

True Acceptance Rate
(@FAR 1E-4)

99.05%

98.95%

98.35%

True Acceptance Rate
(@FAR 1E-5)

99.05%

98.95%

97.03%

True Acceptance Rate
(@FAR 1E-6)

98.89%

97.99%

94.98%

Model Size (MB)

300 MB

17 MB

6.7 MB

Template Size (KB)

5 KB

3 KB

FaceMe^® Technology

Features List

Face Detection

Yes

Face Recognition

Yes

Face Attributes Recognition

Gender, Age, Emotion, Pose

Real-time Video Support

Yes

Emotion Detection

Happy, Sad, Angry, Surprised, Neutral

Images Pre-Processing

Lighting Enhancement, Edge Sharpening, Upsampling

Anti-Spoofing with 3D-Cameras

Yes

Anti-Spoofing with 2D-Cameras

Yes(with or without user interactions)

Anti-Spoofing with Voice

Voice recognition / Speaker verification

Chipsets (CPU/GPU/VPU)

Intel^®64, ARM64^™, NVIDIA^® GPU, NVIDIA^® Jetson, Intel^®Movidius^™ VPU, MediaTek i350 APU, Qualcomm SoC(GPU), NXP i.MX8

Programming Interface

C++, HTTP, C#, Perl

AI Inference Engines

Intel^® OpenVINO^™, NVIDIA^® TensorRT / CUDA, Qualcomm SNPE, MediaTeK NeuroPilot, CoreML

Against Identity Fraud

Yes

e-Passport Support (NFC)

Yes

System Requirements

For Server

Ultra High Precision Model (UH)

Very High Precision Model (VH)
High Precision Model (H)

CPU

Intel^® Xeon^® Processor E3 Family with 4 cores
(E5 Family with 8 cores recommended)

GPU

NVIDIA^® Quadro RTX 4000 or
NVIDIA^® Tesla T4

—

Memory

4GB (16GB recommended)

Windows 10, Windows Server 2019, Ubuntu 16/18

For Mobile Devices

Android ARM64

iOS

CPU

Snapdragon 625
(835 recommended)

Apple A8
(A10 recommended)

Memory

1 GB
(1.5 GB recommended)

Android 5.0

iOS 13

Runtime Memory

500 MB

Performance Numbers

For Server

CPU

AMD EPYC 7502 32-Core

GPU/VPU

NVIDIA^® RTX A6000

FaceMe^® Extraction Model

Execution Time (ms) Detection + Extraction

3.9 ms

2.4 ms

Frames per second (fps)

256.4 fps

416.7 fps

Face Match Time (ms)

0.0018 ms

0.0017 ms

Memory (GB)

3.9 GB

3.7 GB

* 1080p images, 1 face per image

* The "High Precision (DNN)" model is used for Face Detection

* Face Match Time: Time taken to match two templates. For example, given a template extracted from a face image, to identify who is this person in a 1,000 people database takes 1,000 times matching

* FaceMe^® Windows 5.4 is used

For Mobile Devices

Soc

Snapdragon 845

A12X

Device

Google Pixel 3

iPad Pro 12.9" 2018

FaceMe^® Extraction Model

Execution Time (ms)
Detection + Extraction

55.1 ms

40.9 ms

29 ms

22 ms

Frames per second (fps)

18.1 fps

24.4 fps

34.5 fps

45.5 fps

Face Match Time (ms)

0.0006 ms

0.0025 ms

0.0009 ms

Soc

Snapdragon 845

Device

Google Pixel 3

FaceMe^® Extraction Model

Execution Time (ms)
Detection + Extraction

55.1 ms

40.9 ms

Frames per second (fps)

18.1 fps

24.4 fps

Face Match Time (ms)

0.0006 ms

Soc

A12X

Device

iPad Pro 12.9" 2018

FaceMe^® Extraction Model

Execution Time (ms)
Detection + Extraction

29 ms

22 ms

Frames per second (fps)

34.5 fps

45.5 fps

Face Match Time (ms)

0.0025 ms

0.0009 ms

* Tested with Google Pixel3 and iPad Pro

* 720p images, 1 face per image

* The "High Precision (DNN)" model is used for Face Detection

* FaceMe Android 5.3, FaceMe iOS 3.12 are used