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

Axis IR Multi Imager Camera Tested (P3717-PLE) on Feb 18, 2019
Axis has released their first IR multi imager, the P3717-PLE, a repositionable model listing 360° IR illumination and flexible positioning,...
Hanwha Smaller Multi Imager Tested (PNM-9000VQ) on Feb 14, 2019
Hanwha's first repositionable multi imager PNM-9081VQ tested well, but was huge, over 12" wide and weighing in at over 10 pounds. Now, they have...
Bandwidth vs Low Light Shootout - Avigilon, Axis, Bosch, Dahua, Geovision, Hanwha, Hikvision, Uniview, Vivotek on Feb 08, 2019
Nighttime bandwidth spikes are a major concern in video surveillance, but do all manufacturers' cameras perform the same? Are some more consistent...
Sony Favorability Results 2019 on Feb 06, 2019
Sony Favorability amongst integrators improved moderately compared to their 2017 favorability results, with a modest net positive...
Camera Course Winter 2019 - Last Chance on Jan 24, 2019
This is the last chance to register for the Winter 2019 Camera Course. This is the only independent surveillance camera course, based on in-depth...
Testing Bandwidth Vs. Low Light on Jan 16, 2019
Nighttime bandwidth spikes are a major concern in video surveillance. Many calculate bandwidth as a single 24/7 number, but bit rates vary...
WDR Tutorial on Jan 11, 2019
Understanding wide dynamic range (WDR) is critical to capturing high quality images in demanding conditions. However, with no real standards, any...
Worst Products Tested In Past Year on Jan 09, 2019
IPVM has done over 100 tests in the past year. But which products performed the worst? Which ones should users be most aware of? In this report,...
2019 Video Surveillance Cameras Overview on Jan 07, 2019
Each year, IPVM summarizes the main advances and changes for video surveillance cameras, based on our industry-leading testing and...
IPVM Best New Products 2019 Opened - 70+ Entrants on Jan 07, 2019
The inaugural IPVM Best New Product Awards has been opened - the industry's first and only program where the awards are not pay-to-play and the...

Most Recent Industry Reports

From The Basement To Buried Behind Chinatown: ISC West Emerging Technology Zone on Feb 22, 2019
What does ISC West think about 'Emerging Technology'? Well, last year, they put those companies in the basement. This year, they moved them up to...
Private School IT Manager Surveillance Interview on Feb 22, 2019
This IT manager describes himself as the "oft-maligned IT person" whose "opinions may not always be appreciated by the integrator crowd." But he is...
Outdoor Camera Mounting Hardware Guide on Feb 21, 2019
Mounting cameras outdoors can be challenging, requiring understanding different types of equipment and methods. In this guide, we teach this...
HID Favorability Results 2019 on Feb 21, 2019
HID favorability results were strong, in the 2019 IPVM integrator study of 200+ integrators, with a net +62% and low negativity as the table below...
First US State, Vermont, Bans Dahua and Hikvision on Feb 21, 2019
The first US state, Vermont, has issued a ban on a number of Chinese and Russian manufacturers including the world's 2 largest video surveillance...
ADI 'SAVE BIG' On FLIR And Hikvision Examined on Feb 20, 2019
One is a major US defense supplier. The other is owned by the Chinese government. But you can "SAVE BIG" on both at ADI. In this note, we...
BluB0x Company Profile on Feb 20, 2019
BluB0x has doubled in revenue every year since its founding in 2013, according to CEO Patrick Barry. We originally reported on them in 2015. At the...
Massive Leak Of Chinese VMS Provider Exposes Xinjiang Surveillance on Feb 20, 2019
A subsidiary of China’s claimed largest VMS provider is tracking the precise location and ethnicity of millions in China’s Xinjiang region,...
Security Installation Tools Guide - 22 Tools Listed on Feb 19, 2019
In this guide, we cover 22 tools that security installers frequently use. This is one part of our upcoming Video Surveillance...
Sales Cuts At Rasilient on Feb 19, 2019
Over the past 2 years, video surveillance storage specialist Rasilient has expanded its workforce significantly, aiming to build its own branded...

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