This is really cool and really scary at the same time. Sometimes being in this industry is absolutely amazing and sometimes you have to step back and say, Did I really just do that? Referring to making big brother even more powerful!!
Neurala Deep Learning Platform Profile
Published Aug 31, 2017 15:33 PM
Find lost children? If Neurala has its way with its recent partnership with Motorola, that is the aim as shown in their video below:
Neurala has developed a deep-learning platform designed for implementation in low-power edge devices, providing analytics capabilities often found only in more expensive and powerful hardware.
Motorola is deploying Neurala's technology in an updated line of police body cameras, and Neurala hopes to find more security-based uses for their technology.
IPVM spoke with the company, and provides an analysis of their development and potential to impact the security industry.
Background *******
******* *** ******* ** **** ** 3 ****** ********** ***********, ************ ******* [**** no ****** *********] (***) ******* **** ******* [**** no ****** *********] (***) *** ******* ************* [**** no ****** *********] (***). *** ******* ******* out **** **** *******, *** ******, and ***** ****** * ***** ** ~$15M ** ******* **** ****** ******* frameworks ******** *** *** ** ***-***** GPU's. ***** *** *** ** ****** devices **** ******, ******** *******, *** consumer *********** **** ********-****** ************ ********** unavailable ** ***** ********** *** ** size, ****, ** ***** **********.
*** ******* ******* ~** ****** ** its ****** ************, *** ******** ** them ** ***********/*&* *********.
Edge ******** - *** **************
*******'* ********** ****** **** ******* ** be *********** ******* ** ****-**** ** new ****** *******, ********* **** ***** typically ******* **** ******** *********** ********** resources, *** ********** **** ****** (*****). Neurala refers ** **** ********** ** ** LDNN, ** ******** **** ****** *******.
**** *******'* **** ******** ** * video ******, * **** *** ******* video ** *** ******, ********* ** object ** *** *****, *** ***** the ****** ** ********* ****** *********** of **** ******.
**** ********** ****** **** ***** "**** learning" ********* ***** ********* **** ******* understanding ** ******* ******* *******, *** are ****** ** ***** ** ******** new ****** ***** ** ********** ******* the *** ** *********** *********.
Nvidia ************
*** ****-***** ******** *** *************** ** ******'* ****** ***. ******* ** **** **** ********'* ********* *******, ******** ** **** ****** ******* companies **** ******* ****** ******** *** deep ******** ************.
Motorola ****
******** ** ***** *******'* *********, ********* an *********** *** *********'* $*** ****** *** **** ** *** *********** ******* the *** *********. ********'* ***** ******** ******* *** ******** ** **** ********* ** AI/DNN ************, ******** *******'* ******** ** run ****** ***** *******.
Bodycam ************ ** ******* ****
** ********* *******'* ********** ** ********, *** ******* *** ** *********** trained ** ********* ******* ** ******* of ********, ******** *** ******* ******** or ********* **** ** ******* ** appearance. ******** ********* ******** *** ******* "showing" *** ****** ****** ** *** person/object ** ** **** *** ********. A ***** ***** *** ** ****, or * ***** **** *** ** used (**** ** **** ******** ******** footage, ** **** *******), ****** **** accurate ******* *** ******** ** ***** a ****** ** ****** (***** ** video) ******* ** * ****** "*******" style *****. *** ******* **** ********* the ****** *******, ********* ** * period ** * *** *******, ** create ** ******* ******* ** *******, which ** *** **** ********* ** live ***** **** **** ***** *******.
*** ********* ****** *** ****** *** also ** **** ** **** ********** on ******* ** ******* *********** ** an ******* ****** *** ****** ** their *****. **** *** ******* *********** with ********** **** ** *** ***** an ******* ****** *********** ******* *****, groups ** ******, **********, ***** ********, street *******, ***. **** **** *** be **** ** ******** ****** ******, or ***** ***** ****** ****** *** municipality.
Surveillance ************ ** ******* ****
***** ************ ******* *** ********* ** not **** *** **** ********* ********, and *** ********* ** * ********, ***** has more ********** *****, *** **** *** edge-learning ****** * ****** ****** ** reduced. *******, ******** ********* ***** ****** low-cost ***********, *** **** *** ***** resources, ***** *******'* ********** ** ********* could ******* ************ ******* ** ***** listed ** *********'* **** ******** ********,******* ********* * ****** **** **** NVIDIA. **** ** ******-***** ********* **** more *********, *** ******* ** ** dynamic ******** ** * *** **** reduces ******* ************ *** ** ********** to ******** **** ***********.
***** ******* **** ****** ****-***** ********* capabilities *** ****** **************, * ****** of ************* **** ** ********, ****, Hanwha, *********, *** ******, **** ******* begun ********* ******** **** ****-***** *********, making *******'* ******** ** **** ****** of ***** ******** *** *** ************ market.
Upcoming ******* ** **** ******** ****
******* ****** **** **** **** ******* a ******** ****** **** ****** ****** training **** ** *** ****** ** be ******** *** ****** ** ***** devices ****** *** **** ************. **** would ***** ** ***** *** ******* their **** ****** ** *** ********** of * ****** ** ***** **** custom ******** **** **** ***** ********.
Looking *** ******** ******** ********
*** ******* ****** **** ***** **** to ******* **** ******** ******** ************* to ******* *******'* ********** ** ***** products. **** ***** ****** ****** *** appropriate *** ** *** ******, *** then ********* *******'* ********. *** *** requirement ***** **** **** ************* ***** need ** ****** *** ******** ******* for ******** ******* ***** **** ****** to ****** *** **********.
Challenges *** ******* ** ********
***** *******'* ********** *** ** ********** for ************* ** ******** ******* ************, adding even ***-*** *** ************ *** ******** street ****** ** **'* ** *******, which *** ** * *********** ******** of ***** *** **** ******* *** recorders. ************, ***** ************* **** **** ******** to **** ** *** ******** *** core ******** **** ** *********, ********** instead ** ****** ***** *** ******** as * ***** ** ***************. *******, mid-tier ******* *** **** *******'* ********** *** offer **** * *** ** ******* DNN-enabled ******** ******* ****** ** ****** heavily ** ***** *** *&*.
Comments (2)
SD
Shannon Davis
Aug 31, 2017
Skip Cusack
Sep 06, 2017I'm not too concerned yet. The article is very light on technical detail, and knowing Brian's inclination to nail that sort of thing, it's probably because the vendor is light on sharing it. A cop's body camera that beeps away with false alarms is not going to be accepted. And high accuracy is at odds with low cost and small form factor edge devices. It's a great idea, a good vision and deep learning holds the promise of a lightweight footprint, but there is no detail yet. What complexity can it learn? What is the impact of environmental conditions (shadow, low light, direct light, motion, etc.)? What is the accuracy in terms of false reject and false accept? It ain't real until this is all quantified, hopefully by a credible third party.