Amazon Rekognition Facial Detection Tested

Author: IPVM Team, Published on Dec 03, 2018

Amazon's Rekognition service has announced improvements, that if accurate, would be a significant advance over conventional facial detection.

amazon recognition

In particular, Amazon claims:

Amazon Rekognition can now detect 40 percent more faces – that would have been previously missed – in images that have some of the most challenging conditions described earlier... These aspects might include pose variations caused by head movement and/or camera movements, occlusion due to foreground or background objects (such as faces covered by hats, hair, or hands of another person in the foreground), illumination variations (such as low contrast and shadows), bright lighting that leads to washed out faces, low quality and resolution that leads to noisy and blurry faces, and distortion from cameras and lenses themselves. [emphasis added]

IPVM developed a dataset of 100+ 3-second video clips:

facial detection dataset 2

These clips are done across a number of common real-world scenarios:

  • Day light (200 lux)
  • Low light (1 lux)
  • Side tilt (from 0° to 90°)
  • Down tilt (from 0° to 30°)

We compiled various combinations (such as 200 lux, 15° side tilt, 22° down tilt or 1 lux, 75° side tilt, 30° down tilt, etc.).

We then ran tests of these 100+ clips across facial detection approaches:

  • HAAR
  • HOG
  • Openvino (specifically the face-detection-retail-0004 model on Intel i7 and Myriad X chips expanded upon from our NCS2 test)
  • Amazon Rekognition

Inside, we explain full results and key variances in performance.

******'* *********** ******* *** ********* ************, **** ** ********, ***** be * *********** ******* **** ************ ****** *********.

amazon recognition

** **********,****** ******:

****** *********** *** *** ****** ** ******* **** ***** – that ***** **** **** ********** ****** – ** ****** **** have **** ** *** **** *********** ********** ********* *******... ***** aspects ***** *******pose ********** caused by head movement and/or camera movements, ************ ** ********** ** ********** ******* (**** ** ***** ******* by ****, ****, ** ***** ** ******* ****** ** *** foreground),illumination ********** (such as low contrast and shadows), bright lighting that leads to washed out faces, low ******* and resolution that leads to noisy *** ****** *****, and distortion from cameras and lenses themselves. [emphasis added]

**** ********* * ******* ** ***+ *-****** ***** *****:

facial detection dataset 2

***** ***** *** **** ****** * ****** ** ****** ****-***** scenarios:

  • *** ***** (*** ***)
  • *** ***** (* ***)
  • **** **** (**** *° ** **°)
  • **** **** (**** *° ** **°)

** ******** ******* ************ (**** ** *** ***, **° **** tilt, **° **** **** ** * ***, **° **** ****, 30° **** ****, ***.).

** **** *** ***** ** ***** ***+ ***** ****** ****** detection **********:

  • ****
  • ***
  • ******** (************ *** ****-*********-******-**** ***** ** ***** ** *** ****** X ***** ******** ******** *** **** ****)
  • ****** ***********

******, ** ******* **** ******* *** *** ********* ** ***********.

[***************]

Key ********

****** *********** ****** ********* ******** *** *** ****** **** *** others ****** ********* ****** ** **** ** ********:

facial detection overall accuracy

*** ****** *** ********* ***** '***' *** **** ** *** examples ** **** **** ***** ****, ********* *** ********* *** bad ******.

*** *******, **** ** ******* *** ****** ** *** ********** used, **** *** ****** ***********, ***** *********** ** *********** ******* are *******:

rekognition accuracy segmented

**** ******* *** *** ****** ****** ******** ***, **** '****' light *********** ** ****** **% ***** *** ***** *********** ** under **%.

***** *********** *** ********* ** ****** ********* ********, ** ***, far *** ****, *** *******, ** *** ****** ** ********* versus ******** *** **** ******* ** ** **:

facial detection fps

*********** ** **** ****** **** *** ***** ******** ** *** Rekognition ***** ******* ******** *** ***** ** ** **** ***** to *** ***** *** ***** ******** *** ** **** ******* / ** *** **** / ***** *** ****** **. ** examine ***** *********** ******* ** *** ***** ******* ** **** report.

********

*****, ** ******* ** ***% ******** *** ****** ***********, ***** is ****** ****** *********** *** **** ** **° **** *** 45° **** **** **** *** ******:

[video-to-gif output image]

****, ** ******* ** **% ******** (******* ~* *** ** 10 ****** ********* ******* * ****) *** ****** ***********, ****** a ***** ********* ** ***** **** *** **** **** **** the *******'* **** ****:

*** ***** ******, ** *****, **** **** **** *********** **** for ***********. *** *******, **** * *** ** ***** * frames *** * **** ********** ******** ****** **** * ***** angle *** *** ****** / ********:

*******, *** ** *** *** **** ************ *** *********** *** in **** **** ********, ****** *** *****, *** ******* ****** straight ******* *** ****** *** ***** **** *** *** ******** at ***:

Accuracy ***********

*** ******** ****** **** ** ************** *** * **************. *********** face ********* ********** **** * ****** ******"**** ******* *********" (***), ***** ******** ***********-**-******* ** ****** **** ********* ******** ************** ** ****** **** ********-*** ******* (********* ****** "************ **** Union" ** ***) *** ******* **** *** ********* ******** * face. ** ****'* ***** *******, **** **** ****** *** *** face (***** ***** *** * ****, *** ***** ** **** one). ***** ****, ** *** *** * ****** ********* *** accuracy: ** *** **** ****** *** * **** ******** (******** / *****-***** = ********). **** ** *** *******: ** **********, it **** *** ******* *** ***** ********* (* **** ** detected ***********, **** ** *** **** ** ** *** *******'* shirt). **** ******* ****** - ** ** ***** **** *** heuristic. ** ****** ** **** ******* **** ********** (***, *****, etc) ******* ****** ** ***** ******** ***** *** ***** ***** of *** *******.

FPS ***********

*** ******* **** ****** **** *** *********** ***:

  1. ****** *** ***** ** ****** **
  2. ***** *** *****
  3. ***** *********** (******** ** ** *** ** ****, *********** *****->***********)
  4. **** *** *****, ********* ***

* **** ****** ******** *** ************ ***** ** ** ****** videos ******** **********->***********. **** ***** *** *** ********** ******** ** * *****-**-***** basis, ***** *** ***** ***** *** ********** ******** **** *** FPS *********** ********. ** ****, *** *** ***********, ******* **** low, ***** ** **** *****.

***** ** **** ************, **** ** * ***** ***** ******* Rekognition, ** ****** ** *** ** * *******. *** *******, one *** *** ***-***** ******** *** ****** ******* ***********. *** other ****** ** **** ******* *** ** ****** * ** Intel ** ******* (****-***) *******, ** **** *********** *****-*** ** situation-dependent. ************, *********** ** * ******* *******, ***** *********** * model (**** ********'* ****-******-****) ***** **** ************* *** *****. ** while ****-**** *********** **** *********** ** ******** (****: ** *** still ************* ** **** ** ********), *** ***-****-**** ************, *********** could ** ***** ******.

*******

*** * ****-**** ***** ************ ***********,*********** ************ ** ***** **** *** *** ***** ************ ***** **** infrequent ***** ********** ** ******, ** ***** ** *** **** affordable **** **-**** **********. *** *******, *********** ******* *** **-**** (N. ********) ** $*.** *** * *** ** **** ****** video ********. **** **** **** **% ******** ** ******* (*.*., 4,380 ******* ** * *****), *** ******* **** ***** ** ~$50 (*,*** * **% * $*.**) **** *** *** **** analytics.

** *** ***** ****,****** *********** ****** ******* ***************** ****** ***********, ****** ******** (***, ******, *******), ****** *** activities *********, ***. **** **** ********* ******** ****** ** *** have ** ***** ** ****** ** ******** ** ******* **** their ***********.

********

**** ** *** *** ** ****'* *** ****** ** **** Learning *****, ******** ********** ********* ******,***** ****** ******* ***** * / ******** ** ******* *** ****. ** *** ******** * ******* ** *** tests ** *** ******** ******. *** ********, ******** ** *********, please *** ** *** ********.

Comments (7)

*** ***** ** *** ********** ** *** ************,***** *** *** *** ***** **** ****:

deep learning webinar2

***** **** ** ********** ************ ** **** ** *** ******** effort ** ******* ************* ******** *** *********, ** ******** ** our ******* *****.

*** *** ******* **** ******* *** ********** **** ** *********** on? **** ******* ***** ****** ** ** *********** (** **** have *** ******) ** * ***

** *** *** ** *** ** ******** *** ** **** try **** ** ****** *****.

**** **** ***** **** ********* ** **** **** ** **** haven't *******.

** ***** *** *** *** *** **** *** ** **** raw ***** **** ** *** ***** *** **** ***** ** download *** *** **** ***** *** ********* *******?

* ***** **** ***** ** * ****** ***** *** ** compare ********.

** ***** ******* *** ******* *** **** *** ******* ***.

Login to read this IPVM report.
Why do I need to log in?
IPVM conducts unique testing and research funded by member's payments enabling us to offer the most independent, accurate and in-depth information.

Related Reports on Low Light

Genetec Privacy Protector Tested on Nov 12, 2018
Genetec has built Kiwi Security's Privacy Protector into Security Center, an analytic which anonymizes individuals in cameras' fields of view...
Dahua Dual Imager Dome Camera Tested (HDBW4231FN-E2-M) on Nov 07, 2018
Dahua has introduced a dual-imager dome model, the HDBW4231FN-E2-M, with two independently positionable sensors including integrated IR, not found...
Camera Course January 2019 on Nov 01, 2018
This is the only independent surveillance camera course, based on in-depth product and technology testing. Lots of manufacturer training exists...
Axxon Face Search Tested on Oct 26, 2018
AxxonSoft has brought facial recognition to their Axxon Next VMS for free with the simply named Face Search, claiming to allow users to find...
Hanwha Dual Imager Dome Camera Tested (PNM-7000VD) on Oct 18, 2018
Hanwha has introduced their first dual-imager model, the PNM-7000VD, a twin 1080p model featuring independently positionable sensors and a snap-in...
Dahua Face Recognition Camera Tested on Oct 15, 2018
Dahua has been one of the industry's most vocal proponents of the value that AI creates: As part of this, Dahua has released a facial...
Axis Vs. Hikvision IR PTZ Shootout on Sep 20, 2018
Hikvision has their high-end dual-sensor DarkfighterX. Axis has their high-end concealed IR Q6125-LE. Which is better? We bought both and tested...
Axis First IR PTZ Tested (Q6125-LE) on Aug 21, 2018
Axis is very late in releasing IR PTZs. While competitors such as Hikvision and Dahua have offered them for years, Axis has just released their...
Hikvision PanoVu Mini Tested (Multi-imager + PTZ For ~$500) on Aug 07, 2018
Hikvision has released their first PanoVu Mini multi imager, the PanoVu DS-2PT3326IZ-DE3, with four 1080p imagers, including a PTZ and integrated...

Most Recent Industry Reports

The 2019 Video Surveillance Industry Guide on Dec 10, 2018
The 300 page, 2019 Video Surveillance Industry Guide, covers the key events and the future of the video surveillance market, is now available,...
Multi-Factor Access Control Authentication Guide on Dec 10, 2018
Can a stranger use your credentials? One of the oldest problems facing access control is making credentials as easy to use as keys, but restricting...
Top 2019 Trend - AI Video Analytics on Dec 10, 2018
160+ Integrators answered: What do you think the top industry trend will be in 2019? Why? AI / video analytics was the run-away winner with...
AV Tech Company Profile on Dec 07, 2018
Taiwanese manufacturer AV Tech's revenue declined ~70% since 2012. Planning a comeback, AV Tech spoke to IPVM about their opportunities and...
Ubiquiti $79 Flex IP Camera Tested on Dec 07, 2018
U.S. Manufacturer Ubiquiti has released a 1080p, integrated IR IP camera, selling it directly for $79, making this one of the least expensive IP...
Infinova's Xinjiang Business Examined on Dec 07, 2018
As pressure mounts for companies to stop doing business in China’s Xinjiang region amid a severe human rights crisis, IPVM has found Infinova sold...
Akuvox Intercom Profile on Dec 06, 2018
Akuvox, a Chinese manufacturer of VoIP products, is expanding heavily into Video Intercom products with disruptive pricing targeted for commercial...
Sublethal Camera Gun Examined on Dec 06, 2018
Sublethal is a South African company that manufactures a remotely-controlled, camera-enabled gun called the Boomslang, which is Afrikaans for tree...
UK ICO Denies IPVM GDPR Complaint Against IFSEC, Decides Each Exhibitor Responsible on Dec 06, 2018
The UK Information Commissioner's Office (ICO) has denied IPVM's complaint against IFSEC for misuse of facial recognition. Each Exhibitor...
VMS Live Monitoring Shootout - Avigilon, Dahua, Exacq, Genetec, Hikvision, Milestone, Network Optix on Dec 05, 2018
Viewing live video is the first interaction and most common task most users have with a VMS. Who does it best and worst? Who offers the most...

The world's leading video surveillance information source, IPVM provides the best reporting, testing and training for 10,000+ members globally. Dedicated to independent and objective information, we uniquely refuse any and all advertisements, sponsorship and consulting from manufacturers.

About | FAQ | Contact