BRN Discussion Ongoing

[Sirod69]

A Shift in Computer Vision Is Coming​

Prophesee’s evaluation kit for its DVS sensor developed in collaboration with Sony. Benosman is a co-founder of Prophesee. (Source: Prophesee)

General-purpose neuromorphic processors are lagging behind their DVS camera counterparts. Efforts from some of the industry’s biggest players (IBM Truenorth, Intel Loihi) are still works in progress. Benosman said that the right processor with the right sensor would be an unbeatable combination.

https://www.eetimes.eu/a-shift-in-computer-vision-is-coming/

 

[thelittleshort]

Pretty interesting from Amazon

How the Zoox robotaxi predicts everything, everywhere, all at once | Amazon Science | 21 comments

The Zoox autonomous robotaxi can predict the trajectories of vehicles, people, and even animals in its surroundings, as much as 8 seconds into the future. Learn how deep learning, graph neural networks, and convolutional neural networks help inform Zoox’s predictive capabilities...

www.linkedin.com

 

[cosors]

@Filobeddo is still with us, I think he got upset about our iceberg discussion

View attachment 13750

It's interesting how some take a step back and still stay with us. I find that good! Maybe someday the time will come that FF Rise from the ashes.) I begrudge everyone the time out. Yak you are also very welcome from my side. With them there would be already three. But still no Bacon. I find that a pity. So I guess all four of you, no five! Would Filobeddo get his name back if he changed his mind
___
Quite interesting! Only six months (ok I hold longer) and you are already my pack.

 

[Sirod69]

Neuromorphic Device with Low Power Consumption​

Figure 1: The role of RRAM devices in neuromorphic circuits: (a) scanning electron microscopy (SEM) image of an HfO2 1T1R RRAM device, in blue, integrated on 130–nm CMOS technology, with its selector transistor (width of 650 nm) in green; (b) basic building block of the proposed neuromorphic circuit; (c) cumulative density function of the conductance of a population of 16–Kb RRAM devices, as a function of the compliance current ICC, which effectively controls the conductance level; (d) measurement of the circuit in (a); (e) measurement of the circuit in (b). (Source: “Neuromorphic object localization using resistive memories and ultrasonic transducers,” in Nature Communications)

While traditional processing techniques sample the detected signal continuously and perform calculations to extract useful information, the proposed neuromorphic solution calculates asynchronously when useful information arrives, increasing the system’s energy efficiency by up to five orders of magnitude.

https://www.eetimes.com/neuromorphic-device-with-low-power-consumption/

 

[Sirod69]

Neuromorphic Device with Low Power Consumption​

View attachment 13753
Figure 1: The role of RRAM devices in neuromorphic circuits: (a) scanning electron microscopy (SEM) image of an HfO2 1T1R RRAM device, in blue, integrated on 130–nm CMOS technology, with its selector transistor (width of 650 nm) in green; (b) basic building block of the proposed neuromorphic circuit; (c) cumulative density function of the conductance of a population of 16–Kb RRAM devices, as a function of the compliance current ICC, which effectively controls the conductance level; (d) measurement of the circuit in (a); (e) measurement of the circuit in (b). (Source: “Neuromorphic object localization using resistive memories and ultrasonic transducers,” in Nature Communications)

While traditional processing techniques sample the detected signal continuously and perform calculations to extract useful information, the proposed neuromorphic solution calculates asynchronously when useful information arrives, increasing the system’s energy efficiency by up to five orders of magnitude.

https://www.eetimes.com/neuromorphic-device-with-low-power-consumption/

I know is an older article but edge impuls posted it today

Edge Impulse @EdgeImpulse 45min
.
@CEA_Leti researchers are coupling innovative sensors with RRAM–based neuromorphic computation to build ultra-low-power systems for edge AI applications.

 

[cosors]

Credit to @chapman89

Kathrin works in mercedes benz just replied to jesse's comment on linkedin.
View attachment 13729
View attachment 13730

I'm a little distracted. She looks like I think I might like her. I listen to her words of course and do not get distracted

 

[Deleted member 118]

How cool

https://www.bbc.com/news/business-62007675

 

[alwaysgreen]

Pretty interesting from Amazon

View attachment 13752

How the Zoox robotaxi predicts everything, everywhere, all at once | Amazon Science | 21 comments

The Zoox autonomous robotaxi can predict the trajectories of vehicles, people, and even animals in its surroundings, as much as 8 seconds into the future. Learn how deep learning, graph neural networks, and convolutional neural networks help inform Zoox’s predictive capabilities...

www.linkedin.com

I'm no @Diogenese but pretty sure this is not Akida. It's using a CNN not a SNN.

OGRE.GIF?

 

[thelittleshort]

I'm no @Diogenese but pretty sure this is not Akida. It's using a CNN not a SNN.

OGRE.GIF?

@alwaysgreen apologies for confusion. Posted as interesting tech only rather than saying it’s Akida. I’ll delete post if ogred to avoid muddying of waters

 

[Rskiff]

Slightly off topic, but great video by Alex from Ticker Symbol:you on NVIDIA and it's challenges.

 

[equanimous]

CHIPS act signed.
I would be surprised if Brainchip haven't tried to leverage their technology with big names to benefit from this.

Time will tell i guess

You never let us down with chips news Mr Potato.

 

[Foxdog]

All the applications this Renesas ZMOD4410 sensor can be used for is endless.

	Air Conditioner (High-End)
	Air Duct System
	Air Purifier Sensor Module
	Air Quality Control for IoT Building Automation
	Arduino Shield Sensor Board
	Bathroom Odor Detector
	Bathroom Odor Detector with Bluetooth Low Energy
	Biosensing with Wireless Charging and Bluetooth
	Bluetooth Low Energy (LE) Sensor Network Solution
	Building Automation Lighting with Air Quality Sensors Solution
	Cellular Cloud Connected System
	Cloud and Sensor Solution for IoT Endpoints
	Cloud Connected Wi-Fi & Bluetooth® Low Energy Sensor Hub
	Connected Oxygen Concentrator Controller
	Diaper Odor Detector
	Flammable Gas Leakage Detector
	Furnace Control
	Gas Sensor with Cloud Connection for Industrial Applications
	HVAC Air Quality Sensor
	IEEE 802.15.4g-Based Battery-Powered Sub-GHz Wireless Communication
	In-Home Air Quality Monitor System
	Indoor Air Quality Sensor (IAQ)
	Industrial Automation Solution with Industrial Ethernet Module
	Industrial CAN Sensor Network
	Industrial Ether Connectable IoT Sensor Hub
	Infusion Level Monitor Using Capacitive Touch Sensing
	Instrument Panel for Light Electric Vehicles
	IoT Cold Chain Monitoring
	IoT Sensor Board with Machine Learning & Bluetooth® Low Energy
	Large Power BLDC Ceiling Fan with PFC
	Mbed™ Based Image Processing Solution
	Modbus ASCII/RTU Slave Board
	Multi-Purpose Air Quality Sensor Solution
	Multi-Sensor Module for Industrial Ethernet
	Multi-Sensor Platform for ASi-5
	Personal Safety Tracker
	Precision TIG Welding Controller
	Secure Cloud & Sensor Solution
	Smart BLDC Air Cooler
	Smart BLDC Fan with Humidity and Gas Sensors
	Smart Industrial Gas Alarm
	Smart IoT Air Purifier
	Smart Range Hood
	Smart Room Controller with DAB Audio System
	Thermopile CO₂ Detector
	Waterproof Shower Controller
	Wearable Activity Tracker
	Wireless Sensor Hub
	Wireless Sensor Network Solutions

Plenty of medical uses in that list for the 1,000 Eyes to watch out for. Ties in nicely with Rob Telson’s comments last year about involvement in the medical field for BRN. This to me is evidence of significant progress in the background and speaks volumes for why Renesas and others are so keen to keep use of AKIDA a secret. Early adopters of this technology will be years ahead of their respective competitors.
Lack of news from the company is actually a good thing as it shows that BRN respect their customers. This type of loyalty will develop into long lasting relationships that will be profitable for decades to come.
My opinion only - be patient, wait for evidence through revenue each quarter. Maybe we will get sugar hits along the way when customers are in a position to announce their involvement with AKIDA (such as the Mercedes announcement).
Until then, thanks to those who are so good at joining the dots

 

[Slymeat]

We already know the EQXX was using Akida IP. If Jesse is our Chapman, he should know. Why is he asking?

He may just be giving Mercedes an opportunity to officially-unofficially reiterate that they use Akida technology. Daily tweets from Mercedes would be perceived as spam, but they have a right, even an obligation, to answer every legitimate question. I don’t see any harm in that, it helps push Akida’s existence to the Twitterverse where a VERY large number of influential people lurk.

 

[TopCat]

Synaptics we’re discussed here back in May and dismissed but this is the NEW Katana Edge kit. Can anyone explain why it shouldn’t be considered?

Edge AI and sensor fusion are two disciplines of application development for engineers due to their complexity and multidimensionality. It is more convenient to have a complete solution that integrates a wide range of sensors with edge devices supporting multiple use cases.

The new Katana Edge AI Evaluation Kit (EVK), built around the low-power Katana SoC (System-on-Chip) with its sophisticated neural network engine, integrates various sensors, software, and wireless connection to jumpstart smart edge designs. Katana has a hexa-core architecture that is heterogeneous and each core is well-suited to application-specific activities for building target use cases. The activities include neural network image processing, audio/voice processing, control tasks, and many more. The Edge AI kit aims to accelerate the Development of AI Vision and Sensor Fusion Applications for the IoT networks.

The Katana platform from Synaptics extends intelligence to the edge by allowing energy-efficient AI endpoint solutions that can make autonomous choices based on data acquired through vision and audio sensors. It offers extremely minimal power per inference in a modular, user-friendly AI architecture. Katana is a high-integration edge AI SoC for battery-powered wireless devices with audio or video capabilities in consumer and industrial IoT applications.

Katana Edge AI Kit Architects Smart Edge Designs for AI Vision and Sensors - Embedded Computing Design

Edge AI and sensor fusion are two disciplines of application development for engineers due to their complexity and multidimensionality. It is more convenient to have a complete solution that integrates a wide range of sensors with edge devices supporting multiple use cases.

embeddedcomputing.com

 

[Bigal7425]

Pretty interesting from Amazon

View attachment 13752

How the Zoox robotaxi predicts everything, everywhere, all at once | Amazon Science | 21 comments

The Zoox autonomous robotaxi can predict the trajectories of vehicles, people, and even animals in its surroundings, as much as 8 seconds into the future. Learn how deep learning, graph neural networks, and convolutional neural networks help inform Zoox’s predictive capabilities...

www.linkedin.com

Hi @thelittleshort
A good friend of mine had been an early private investor in Zoox a few years back before Amazon came in and bought them out for what he thought was sense on the dollar.
He's a pretty major investor, particularly in tech and I have tried relatively unsuccessfully to sell him the BrainChip story for at least two years now (I have been in it for 7 ). He's an interesting study for me, as he like many of the big end, need to see revenue numbers and pathways etc before jumping in. Just makes the mind boggle as to the action we're going to get share wise when those kind of investors boxes are ticked.
I would soooo love to see BrainChip integral and leading in the next developments of autonomous driving as much for a very maturely delivered "I told you so" I to him, as my own personal wealth5
Thanks as always to the ever amazing updates and insights from you all

 

[buena suerte :-)]

Seems like that was actually his last post because his avatar is deleted. I didn't catch what or who he was so upset about. I must have missed something or not understood

I think it shows him as a 'guest' now .... would be good to get all who have left recently to return and carry on contributing the facts/fun/and banter on this great platform. Cheers Chippers

 

[Esq.111]

Good Morning Chippers,

See our mysterious price suppressor is up early...

Sell order pressently sitting at $1.29 for 700,515 units.

Would love to see this order smacked clean out of the ballpark.
Would definitely wipe the grin off this muppets face.

Regards,
Esq.

 

[Deleted member 118]

The Nasdaq has been down the last 2 trading days yet we have been up, so I’m expecting us to take out the resistance around the $1.20 mark today and finish around $1.21-$12.56.

 

[Deleted member 118]

Good Morning Chippers,

See our mysterious price suppressor is up early...

Sell order pressently sitting at $1.29 for 700,515 units.

Regards,
Esq.

That will be shifted a few times today.

 

[Stable Genius]

Live feedback from Washington DC | Synopsys Inc | 56 comments

We are excited to share a moment with our Chairman & CEO, Aart de Geus, live from the White House. Hear his thoughts on this week's signing of the #CHIPSAct. | 56 comments on LinkedIn

www.linkedin.com

If the link works it’s worth a 1 minute listen to Synopsys CEO discuss the mood after a meeting at the White House!

Maybe that’s where PVDM is?