BRN Discussion Ongoing

[Bravo]

Interesting like from Mr Brightfield imo

In January 2019, I was given the honor and challenge to lead the newly formed Hardware team at what would later become Meta Reality Labs. From inception, the team's most ambitious project was Orion… | Rafa Camargo | 39 comments

In January 2019, I was given the honor and challenge to lead the newly formed Hardware team at what would later become Meta Reality Labs. From inception, the team's most ambitious project was Orion, involving close collaboration with the Research and XR Tech teams. We had to build the teams...

www.linkedin.com

How Meta Made Silicon Carbide Waveguides & Unlocked Orion’s Large Field of View

As clear as the choice of silicon carbide as a substrate seems today, when we first started down the road to AR glasses a decade ago, it was anything but.

www.meta.com

Specialized Silicon: How Meta’s Custom Chips Are Revolutionizing Augmented Reality

In 2017, Reality Labs Chief Scientist Michael Abrash, backed by Meta Founder & CEO Mark Zuckerberg, established a new, secret team within what was then Oculus Research to build the foundation of the next computing platform. Their herculean task: Create a custom silicon solution that could...

www.meta.com

Orion’s Compute Puck: The Story Behind the Device that Helped Make Meta’s AR Glasses Possible

Last year at Connect, we unveiled Orion—our first true pair of AR glasses. Today, we’re turning our attention to an unsung hero of Orion: the compute puck.

www.meta.com

View attachment 79054 View attachment 79055

Remember the podcast in January when Steve mentioned he was in ongoing discussions with manufacturers of smart glasses? He said they are working with manufacturers to integrate more advanced algorithms into products.

 

[Bravo]

Speaking of smart glasses, what have we here?

Fortiss is actively developing smart glasses prototypes incorporating neuromorphic technology through their EMMANÜELA project, which focuses on energy-efficient human-machine interaction in augmented and virtual reality (AR/VR) settings. This initiative leverages neuromorphic sensor technology and computing to enhance immersive digital experiences.

Can't find anything to suggest a direct collaboration between Fortiss and BrainChip at this time, so will keep it on the watch list.

Neuromorphic computing creates an immersive digital experience

In the EMMANÜELA project, fortiss is researching the use of neuromorphic sensor technology and neuromorphic computing in AR/VR. The aim is to investigate the potential of these technology-inspired approaches for augmented digital worlds and to open up new application possibilities.

www.fortiss.org

 

[TECH]

1:45 PM - 2:15 PM
Information
Dr. Anthony Lewis
Brainchip

Fast Online Recognition of Gestures using Hardware Efficient Spatiotemporal Convolutional Networks via Codesign​

The Temporal Neural Networks (TENNs) developed by Brainchip can be used to tackle a wide range of vision tasks, including object detection, eye tracking and gesture recognition. Here, we will show how the codesign of model architecture, training pipeline and hardware implementation can combine to achieve SOTA performance, using a gesture recognition task example.
The TENNs architecture leverages multiple techniques to improve its efficiency on compatible hardware (such as the Akida chip). First, although effectively offering a 3D convolution, it uses spatially and temporally separable convolutions to make the model lighter in parameter count with equivalent computational power. Second, when deployed on dedicated hardware, temporal inputs are buffered efficiently to minimize memory usage and data movement. Finally, it is possible to reduce model computation even further by adding regularization to boost sparsity of information transiting in the already slim network (achieving more than 90% average activation sparsity in some layers) and thus further improve the efficiency on compatible hardware.
We apply a lightweight TENN model to a gesture recognition task, showing that it can accurately classify the movements performed by a range of actors with SOTA accuracy. The efficiency of the model is then pushed further with virtually no cost to accuracy by applying regularization of activations.

Looks like Tony will be taking the stage and delivering some magic on behalf of the Brainchip Family !!

In just a little over 7.5 hours from now, so work out your own time zones if you're interested, I am happy he has attended this event,
as our CTO you must respect the fact, yes it's a team thing, BUT Tony is now our leader on the technical side, as Peter has retired, as such.

Comon Brainchip, we are worth way more than our current ASX share price indicates, it's an absolute joke.......Tech x

 

[7für7]

1:45 PM - 2:15 PM
Information
Dr. Anthony Lewis
Brainchip

Fast Online Recognition of Gestures using Hardware Efficient Spatiotemporal Convolutional Networks via Codesign​

The Temporal Neural Networks (TENNs) developed by Brainchip can be used to tackle a wide range of vision tasks, including object detection, eye tracking and gesture recognition. Here, we will show how the codesign of model architecture, training pipeline and hardware implementation can combine to achieve SOTA performance, using a gesture recognition task example.
The TENNs architecture leverages multiple techniques to improve its efficiency on compatible hardware (such as the Akida chip). First, although effectively offering a 3D convolution, it uses spatially and temporally separable convolutions to make the model lighter in parameter count with equivalent computational power. Second, when deployed on dedicated hardware, temporal inputs are buffered efficiently to minimize memory usage and data movement. Finally, it is possible to reduce model computation even further by adding regularization to boost sparsity of information transiting in the already slim network (achieving more than 90% average activation sparsity in some layers) and thus further improve the efficiency on compatible hardware.
We apply a lightweight TENN model to a gesture recognition task, showing that it can accurately classify the movements performed by a range of actors with SOTA accuracy. The efficiency of the model is then pushed further with virtually no cost to accuracy by applying regularization of activations.

Looks like Tony will be taking the stage and delivering some magic on behalf of the Brainchip Family !!

In just a little over 7.5 hours from now, so work out your own time zones if you're interested, I am happy he has attended this event,
as our CTO you must respect the fact, yes it's a team thing, BUT Tony is now our leader on the technical side, as Peter has retired, as such.

Comon Brainchip, we are worth way more than our current ASX share price indicates, it's an absolute joke.......Tech x

YEAHHHH YOUR STATEMENT IS GENIUS LEVEL THINKING … GOOO BRAINCHIIIIP

 

[toasty]

Interesting we've closed up while the index is down quite a bit...........and the buy/sell ratio looks very positive..........

 

[Labsy]

Speaking of smart glasses, what have we here?

Fortiss is actively developing smart glasses prototypes incorporating neuromorphic technology through their EMMANÜELA project, which focuses on energy-efficient human-machine interaction in augmented and virtual reality (AR/VR) settings. This initiative leverages neuromorphic sensor technology and computing to enhance immersive digital experiences.

Can't find anything to suggest a direct collaboration between Fortiss and BrainChip at this time, so will keep it on the watch list.


View attachment 79058



View attachment 79059

Neuromorphic computing creates an immersive digital experience

In the EMMANÜELA project, fortiss is researching the use of neuromorphic sensor technology and neuromorphic computing in AR/VR. The aim is to investigate the potential of these technology-inspired approaches for augmented digital worlds and to open up new application possibilities.

www.fortiss.org

Great find.... If nothing else, you would hate to be developing the latest Meta glasses and watching your competition advertising that they are implementing the latest cutting edge neuromorophic tech, regardless of who's tech it is, and you are not... It's a win.

 

[Rach2512]

A couple of late ones, trying to get in before tomorrow's news!

 

[Luppo71]

Speaking of smart glasses, what have we here?

Fortiss is actively developing smart glasses prototypes incorporating neuromorphic technology through their EMMANÜELA project, which focuses on energy-efficient human-machine interaction in augmented and virtual reality (AR/VR) settings. This initiative leverages neuromorphic sensor technology and computing to enhance immersive digital experiences.

Can't find anything to suggest a direct collaboration between Fortiss and BrainChip at this time, so will keep it on the watch list.


View attachment 79058



View attachment 79059

Neuromorphic computing creates an immersive digital experience

In the EMMANÜELA project, fortiss is researching the use of neuromorphic sensor technology and neuromorphic computing in AR/VR. The aim is to investigate the potential of these technology-inspired approaches for augmented digital worlds and to open up new application possibilities.

www.fortiss.org

Same company using Loihi for camera and many other things.

ELEANOR​

Following on from the INRC3 project, where a robotic arm is taught to insert an object using only force feedback, the ELEANOR project (Energy and Latency Efficient Object Insertion Using a Robotic Arm Equipped with Event-Based KAmera and NeuromOrpher Hardware) uses an event-based camera to make the arm approach the slot. Optical flow and 3D reconstruction via Intel's Loihi research chip is used for this purpose.

IBM and Loihi mentioned everywhere on thier website but no Brainchip or Akida.
Lets hope they are trying something new.

 

[7für7]

Same company using Loihi for camera and many other things.

ELEANOR​

Following on from the INRC3 project, where a robotic arm is taught to insert an object using only force feedback, the ELEANOR project (Energy and Latency Efficient Object Insertion Using a Robotic Arm Equipped with Event-Based KAmera and NeuromOrpher Hardware) uses an event-based camera to make the arm approach the slot. Optical flow and 3D reconstruction via Intel's Loihi research chip is used for this purpose.

IBM and Loihi mentioned everywhere on thier website but no Brainchip or Akida.
Lets hope they are trying something new.

BrainChip is as we know involved at the intel accelerator program… so… who knows… everything is possible. But we will as always just speculate…

 

[GStocks123]

Has this been posted previously? Research from University of Luxembourg. If not, need help digging..

TelecomAI - Lab

The TelecomAI at the University of Luxembourg is used for research and testing of signal processing techniques for satellite and terrestrial communication systems.

www.uni.lu

https://digital-library.theiet.org/doi/10.1049/icp.2024.4619#tab-contributors

 

[7für7]

Has this been posted previously? Research from University of Luxembourg. If not, need help digging..

TelecomAI - Lab

The TelecomAI at the University of Luxembourg is used for research and testing of signal processing techniques for satellite and terrestrial communication systems.

www.uni.lu

https://digital-library.theiet.org/doi/10.1049/icp.2024.4619#tab-contributors

I see a lot of BrainChip papers involved in various projects and forms right? …….


…….BUT I SEE NO MONEEEEEY

 

[Bravo]

Great to see how prominently we're featured in the Wevolver article.

Did anyone notice the header on page 64 mentions 6G as well as discussing BrainChip's AKIDA and TENNs technology?

I guess it's purely coincidental that Qualcomm has decided to acquire Edge Impulse (who have been very supportive of BrainChip's technology) at the same time that they're working towards developing 6G technology.

Here's what ChatGPT had to say in regards to a potential Qualcomm/ BrainChip collaboration to advance 6G. Obviously I don't need to remind everyone to take the ChatGPT response with a grain of salt and to conduct your own research beyond that.

Having said that, it's interesting to note that BrainChip is clearly involved in 6G development, as can be evidenced by this IET research paper on Akida’s role in 5G/6G Non-Terrestrial Networks (NTN).

 

[Bravo]

Has this been posted previously? Research from University of Luxembourg. If not, need help digging..

TelecomAI - Lab

The TelecomAI at the University of Luxembourg is used for research and testing of signal processing techniques for satellite and terrestrial communication systems.

www.uni.lu

https://digital-library.theiet.org/doi/10.1049/icp.2024.4619#tab-contributors

Hi @GStocks123,

Yes, I believe Frangipani first posted on this in early Feb.

• #96,039

 

[GStocks123]

Even this…?

 

[7für7]

One day we will laugh about those days think about that

 

[Frangipani]

The latest Brains & Machines podcast (including transcript) is out!

https://www.eetimes.com/podcasts/brainchips-ip-for-targeting-ai-applications-at-the-edge/

View attachment 77400

View attachment 77399

The articles and papers referred to in the podcast can be accessed on the Brains & Machines website:

Episode 24: Brainchip – Brains and Machines

brainsandmachines.net

View attachment 77401
View attachment 77402

Today, the people behind the Brains & Machines podcast posted a reminder about their recent episode on BrainChip (released on February 7), which is now available on multiple podcast platforms:

Hope you didn't miss our episode on BrainChip and their AI-at-the-edge technology: now available on Spotify, Apple podcasts, and via many other providers. https://lnkd.in/ehmi9jVT | Brains and Machines

Hope you didn't miss our episode on BrainChip and their AI-at-the-edge technology: now available on Spotify, Apple podcasts, and via many other providers. https://lnkd.in/ehmi9jVT

www.linkedin.com

 

[Frangipani]

Even this…?

View attachment 79092

Yup, even this… Sorry!

Here:

https://thestockexchange.com.au/threads/brn-discussion-ongoing.1/post-443061

… as well as here:


https://thestockexchange.com.au/threads/brn-discussion-ongoing.1/post-445787

And @Quiltman posted about that paper, too, when Swetha Varadarajulu first mentioned it in her LinkedIn post:

https://thestockexchange.com.au/threads/brn-discussion-ongoing.1/post-443098

However, no one has come forward to date who’d be able to access the full paper beyond the abstract.

 

[DingoBorat]

Today, the people behind the Brains & Machines podcast posted a reminder about their recent episode on BrainChip (released on February 7), which is now available on multiple podcast platforms:

Hope you didn't miss our episode on BrainChip and their AI-at-the-edge technology: now available on Spotify, Apple podcasts, and via many other providers. https://lnkd.in/ehmi9jVT | Brains and Machines

Hope you didn't miss our episode on BrainChip and their AI-at-the-edge technology: now available on Spotify, Apple podcasts, and via many other providers. https://lnkd.in/ehmi9jVT

www.linkedin.com

View attachment 79113

They have 86 followers..
Let's hope there's a Big Dog amongst them..

 

[Frangipani]

The Akida Edge AI Box User Guide can now be found on Manuals+:

Akida Edge Al Box User Guide

Discover how to set up and configure the Akida Edge AI Box with this comprehensive user manual. Learn how to connect IP cameras, configure network settings, verify connectivity, and find RTSP URLs. Get started easily with the Quick Start Guide included.

manuals.plus

 

[Pom down under]

1:45 PM - 2:15 PM
Information
Dr. Anthony Lewis
Brainchip

Fast Online Recognition of Gestures using Hardware Efficient Spatiotemporal Convolutional Networks via Codesign​

The Temporal Neural Networks (TENNs) developed by Brainchip can be used to tackle a wide range of vision tasks, including object detection, eye tracking and gesture recognition. Here, we will show how the codesign of model architecture, training pipeline and hardware implementation can combine to achieve SOTA performance, using a gesture recognition task example.
The TENNs architecture leverages multiple techniques to improve its efficiency on compatible hardware (such as the Akida chip). First, although effectively offering a 3D convolution, it uses spatially and temporally separable convolutions to make the model lighter in parameter count with equivalent computational power. Second, when deployed on dedicated hardware, temporal inputs are buffered efficiently to minimize memory usage and data movement. Finally, it is possible to reduce model computation even further by adding regularization to boost sparsity of information transiting in the already slim network (achieving more than 90% average activation sparsity in some layers) and thus further improve the efficiency on compatible hardware.
We apply a lightweight TENN model to a gesture recognition task, showing that it can accurately classify the movements performed by a range of actors with SOTA accuracy. The efficiency of the model is then pushed further with virtually no cost to accuracy by applying regularization of activations.

Looks like Tony will be taking the stage and delivering some magic on behalf of the Brainchip Family !!

In just a little over 7.5 hours from now, so work out your own time zones if you're interested, I am happy he has attended this event,
as our CTO you must respect the fact, yes it's a team thing, BUT Tony is now our leader on the technical side, as Peter has retired, as such.

Comon Brainchip, we are worth way more than our current ASX share price indicates, it's an absolute joke.......Tech x

No more news about this?