TensorRT Optimizations for Embedded Facial Recognition Alexey - - PowerPoint PPT Presentation
TensorRT Optimizations for Embedded Facial Recognition Alexey Kadeishvili, CTO, Vocord Vocord Company: Main Facts Developer of video surveillance and video analytics systems since 1999 Deep expertise in facial recognition
Alexey Kadeishvili, CTO, Vocord
www.vocord.com 2
■ Developer of video surveillance and video analytics systems since 1999 ■ Deep expertise in facial recognition ■ Top-rated in NIST and Megaface face recognition tests ■ NVIDIA Metropolis program member Our customers and partners
250+ projects for public and private sectors 140 million faces in enrollment database in a single project 200,000 cameras are managed by VOCORD video analysis software 350,000/month API request to VOCORD FaceMatica cloud Geography: Europe, Middle East, SE Asia, East Asia, Latin America, Oceania
www.vocord.com 3
www.vocord.com 4
All products support NVIDIA GPU
VOCORD FaceMatica
Face recognition engine in a Cloud
VOCORD NetCam
New generation face recognition camera
VOCORD NanoFace
NVIDIA Jetson-based embedded face recognition solution
VOCORD FaceControl
“Faces in the crowd” FR system
VOCORD FaceControl 3D
Free flow 3D facial recognition
nano
Face Recognition SDK
Face recognition engine SDK
www.vocord.com 5
Enrolment DB Recognition engine Inbound image quality
Enrolment DB quality: something beyond control Recognition engine: already works as in the Marvel movies
www.vocord.com 6
TOP in Megaface Face Scrub Open Challenge 2015-2018
With accuracy 91.76%
TOP in NIST Face Recognition Vendor Test 2016-2018
TPR at FPR 10-4 = 98.7%, TPR at FPR 10-6 = 96.6%
www.vocord.com 7
Source: NIST Face recognition vendor test, 2018
www.vocord.com 8
< 10˚ 10 ÷ 30˚ 30 ÷ 45˚ 45 ÷ 60˚ > 60˚ > 60˚, enrollment DB >60˚
0.25 0.2 0.15 0.1 0.05
FRR FAR
1.E-01 1.E-04 1.E-05 1.E-03 1.E-07 1.E-06 1.E-02 1.E00
Enrollment DB <30˚ Group 1 <10˚ Group 2 10 ÷ 30˚ Group 3 30 ÷ 45˚ Group 4 45 ÷ 60˚ Group 5 > 60˚
www.vocord.com 9
L=48 pix L =24 pix
*L – the distance between eyes, pix ** FAR=10-4
Face identification probability Pixels between eyes (L)
0.7 0.8 0.85 0.75 1.0 0.95 0.9 72 48 36 60 12 24
True Identification Rate**
Optimal resolution Recommended minimum
www.vocord.com 10
Recognition engine: already works as in the Marvel movies Enrollment DB quality: something beyond control The quality of acquired face images: point of growth
Enrollment DB Recognition Engine Inbound Image Quality
www.vocord.com 11
NIST FRVT Report 2017 10 03
www.vocord.com 12
Algorithm A Algorithm B
NIST FRVT Report 2017 10 03
www.vocord.com 13
Algorithm A Algorithm B
NIST FRVT Report 2017 10 03
www.vocord.com 14
0.1 0.3 0.4 0.2 0.7 0.6 0.5
FAR FRR
1.E-04 1.E-05 1.E-03 1.E-07 1.E-06 1.E-02
Algorithm A, uncontrolled environment Algorithm B, uncontrolled environment Algorithm A, controlled environment Algorithm B, controlled environment
www.vocord.com 15
0.1 0.3 0.4 0.2 0.7 0.6 0.5
FAR FRR
1.E-04 1.E-05 1.E-03 1.E-07 1.E-06 1.E-02
Algorithm A, uncontrolled environment Algorithm B, uncontrolled environment Algorithm A, controlled environment Algorithm B, controlled environment
www.vocord.com 17
■ Face recognition onboard ■ No compression artifacts: the image is taken directly from the sensor ■ Dynamic Region of Interest for every intelligent algorithm ■ Algorithm adjustment for particular camera set up VOCORD NetCam.AI edge video analytics camera
18
12 bit image with static ROI 12 bit image with dynamic ROI Backlight, no enhancement
Dynamic ROI enhances the quality of image in the face area
www.vocord.com 19
Automated lens control High quality sensor NVIDIA Jetson TX1 GPU
www.vocord.com 20 Camera specs Resolution 3÷5 Mpix Temperature range
Ingress Protection IP 67 Dimensions 20x71x150 mm Power consumption 15W Built-in facial recognition engine specs Min face resolution for face recognition 12 pixels between the eyes Number of faces detected in one frame Up to 25 Latency of biometric template extraction Up to 150 ms per 1 face Face recognition performance Up to 32 faces/s Inference framework TensorRT
www.vocord.com 21 32 19 12 9 6 4 2,2 1,4 0,9 5 10 15 20 25 30 35 "Shallow" CNN "Medium" CNN "Deep" CNN NVIDIA Jetson TX1 Intel Movidius Qualcom Snapdragon 820
www.vocord.com 22
1.E-02
FRR
0.03 0.7 0.15 0.11 1.E-04 1.E-05 1.E-03 1.E-07 1.E-06
FAR
0.13 0.09 0.5 0.01
”Shallow” CNN “Medium” CNN “Deep” CNN Single face: Track (multiple faces): ”Shallow” CNN “Medium” CNN “Deep” CNN
www.vocord.com 23
“Shallow” CNN “Very” CNN “Medium” CNN
Platform: NVIDIA Jetson TX1
FPS TensoRT MXNet
15 35 25 30 20 5 10 32 18 19 10 12 6
www.vocord.com 24
Edge analytics system with VOCORD NetCam.AI cameras 25 “Traditional” server architecture approach with regular IP-cameras
LAN, Wi-Fi LAN One archive server Data center with many expensive rack servers
www.vocord.com 26
Edge computing with VOCORD NetCam.AI
26
“Traditional” server architecture with IP cameras
Cameras USD 2,000 x 100 = USD 200,000 Server for matching and archive USD 10,000 Cameras USD 500 x 100 = USD 50,000 Servers Detection: 2 servers, 4xCPU 32 cores each USD 60,000 Template extraction: 4 servers, 2 GPU Tesla P40 each USD 120,000 Server for matching and archive USD 10,000
CAPEX: USD 210,000 CAPEX: USD 240,000
Maintenance costs: power supply (7-8 kWt), bandwidth (2Gbps), rack space
OPEX: USD 30,000 per year
Maintenance costs: power supply (800 Wt), bandwidth (2Gbps), rack space
OPEX: USD 2,000 per year
www.vocord.com 27
www.vocord.com 28 Facial recognition Behavioral analysis License plate recognition Emergency cases Lost and found objects Vehicle types
www.vocord.com 29 Step 1. The camera collects images and uploads them to the server Step 2. The neural network is retrained on the server using new images Step 3. Customized, light-weight neural network is uploaded back to the camera
www.vocord.com 30 Deeper DNNs provide better performance on unrestricted data On restricted data difference between deep and shallow network is negligible
Unrestricted data Restricted data
0.01 0.015 0.005 0.04 0.025 0.02 0.035 0.03 1.E-01 1.E-04 1.E-05 1.E-03 1.E-07 1.E-06 1.E-02
“Deep” neural network “Shallow” nueral network FAR FRR
1.E-02
FRR
0.01 0.015 0.005 0.04 0.025 0.02 0.035 0.03 1.E-04 1.E-05 1.E-03 1.E-07 1.E-06
FAR “Deep” neural network “Shallow” neural network
www.vocord.com 31
NetCam.AI #1 NetCam.AI #2
Face Jeans Bag
www.vocord.com 32
E-mail: sales@vocord.com Website: www.vocord.com