The next focusIR Investor Webinar takes places on 14th May with guest speakers from WS Blue Whale Growth Fund, Taseko Mines, Kavango Resources and CQS Natural Resources fund. Please register here.
Sorry, this isn't for us
1/14.46" CMOS 160 Kpixel (400 x 400) Image Sensor with PureCel®Plus-S, Global Shutter, and Nyxel® Technologies.
Tiny camera, but only 400x 400 pixels. This will be used inside an AR/VR headset to watch eyes, from extremely close up so that the eye will fill the image. Where as we need mega pixel resolution so that we can get enough pixels across the eye from arms length while also capturing the cabin.
These can also be used on the outside of the headset to pick up the fixed IR "lighthouses" so that they can track movement in 3d and head rotation
Tobii
Tobii CEO Anand Srivatsa headshot
Anand Srivatsa took over as the CEO of Tobii in December. Tobii
Founded: 2001
Headquarters: Danderyd Municipality, Sweden
Tobii, a Swedish eye-tracking firm, raised $121.8 million in funding over five rounds before its initial public offering in 2016. It makes eye-tracking and "attention computing solutions" for the healthcare, gaming, and education industries as well as the auto sector.
In August 2021, Tobii announced it was acquiring Phasya, whose software for analyzing biometric signals helped it launch a driver-monitoring system.
Tobii's CEO, Anand Srivatsa, took the helm in December after serving as a division chief executive officer for another Tobii unit. An Intel veteran, Srivatsa has a master's in electrical engineering from Stanford.
Jungo
Jungo's driver monitoring system
Jungo says its VuDrive system meets the requirements set out by Europe's safety regulators. Jungo
Founded: 2013
Headquarters: Netanya, Israel
A Cisco spinoff, Jungo went public on the Tel Aviv Stock Exchange in July 2021. It works with automakers, major industry suppliers, fleet operators, and aftermarket manufacturers. It's raised $17.5 million in funding over five rounds, the most recent of which was in 2006, according to Crunchbase. Its investors include Cisco and Intel's Communications Fund.
Jungo says its VuDrive system meets the requirements set out by Europe's safety regulators. The product is designed to track a driver's head position and gaze to look for distraction or drowsiness, and the company says fleet operators can use it to "score" their drivers.
The company is run by Opher Suhami, a Cisco veteran.
Seeing Machines
Seeing Machines CEO Paul McGlone headshot
Paul McGlone, the CEO of Seeing Machines, has a background in logistics and supply-chain management. Seeing Machines
Founded: 2000
Headquarters: Canberra, Australia
Seeing Machines, which was spun out from the Australian National University in 2000, provides hardware and software to monitor drivers, pilots, air-traffic controllers, and others. It works with transportation companies including Daimler, GM, Magna, Emirates, and Qantas.
The company launched its Guardian system for commercial trucks and buses in 2016 — it says the driver-monitoring system has since covered more than 800 million miles and detected more than 3.6 million "fatigue and distraction-related driver events." It claims its technology is "scientifically proven to reduce fatigue events by more than 90%."
Seeing Machines has raised $5.7 million in funding, according to Crunchbase. It has a partnership with the chipmaker Qualcomm.
Its CEO, Paul McGlone, has a background in logistics and supply-chain management.
Cipia
Cipia CEO Yehuda Holtzman headshot
Cipia is helmed by Yehuda Holtzman, who took over in May and previously served as the CEO of On Track Innovations. Cipia
Founded: 2005
Headquarters: Tel Aviv, Israel
The Israeli outfit was founded as Eyesight Technologies and originally focused on computer vision. It updated its name in 2020 to reflect its expansion into neural networks — "Cipia" is a reference to the brain's occipital lobe, which turns visual inputs into environmental understanding.
The company says that its products are now in millions of devices and that it has deals with five automakers. It recently announced that it was working with the chipmakers Ambarella and Mobileye and that its driver-monitoring system would be installed in a new electric SUV produced by the Chinese automaker Chery.
Cipia is helmed by Yehuda Holtzman, who took over in May. Holtzman previously served as the CEO of On Track Innovations, a near-field-communication and cashless-payment firm. The company has raised $45.9 million in funding over four rounds, according to Crunchbase.
oops a few bits of code got edited and I can read without signing in ;-)
Driver-watching cameras are expected to jump by 500% to become a $1 billion business. These 4 under-the-radar companies are already conquering the booming market.
Alex Davies 2 hours ago
Depiction of how Cipia's driver-monitoring system watches a male driver's face
Driver-monitoring systems like Cipia's look for signs of inattention or drowsiness. Cipia
Driver-facing cameras will soon be installed in millions of cars thanks to new rules in Europe.
The cameras offer safety benefits as well as new options for facial-recognition features.
They can add $200 to a vehicle's cost, creating an opportunity for suppliers to corner a new market.
Until recently, the handful of companies that supplied the auto industry with driver-facing cameras filled orders of tens of thousands of units a year. Now they're starting to sell millions.
The boom is thanks to a couple of updated regulations across the pond. Starting next year, the tech will be a prerequisite for getting a five-star safety rating under Europe's new-car-assessment program. And come 2024, the European Commission will make it mandatory in all new cars.
Driver-facing camera systems are designed to detect signs of driver inattention or drowsiness and can even automatically slow the car down, said Maite Bezerra, an industry analyst at ABI Research. In a report on the effect of the new rules, she predicted that shipments of driver-monitoring systems would rise by 487.5% between this year and 2027, good for $954 million in revenue.
As they become required to install tech that can add $200 to the cost of a vehicle, automakers are looking at additional use cases to capitalize on the investment.
Facial identification could allow a car to recognize its driver and adjust the seat position, infotainment settings, and more. Knowing a driver's head position and tracking their gaze could allow for more effective augmented-reality applications. Down the road, Bezerra said, it could even help power an AI-based personal assistant.
Bezerra predicted that from 2022 to 2025, shipments of vehicles with driver-monitoring systems would climb from 8 million to 30 million. By 2027, they could hit 47 million and account for more than half of global sales.
Europe would make up a major chunk of those, as would China, which requires the tech in many commercial vehicles. In the US, the National Highway Traffic Safety Administration is talking about adding the systems to its safety-rating system by 2026.
That all spells boom times for the handful of industry suppliers that offer these systems. "The opportunity for them is enormous," Bezerra said. She added that while more players will want in, these four established companies are poised to cash in.
I always get confused reading about head mounted displays. How can my eyes move and my head move and my whole body (in an aircraft?) move, yet despite all of this, it can still compute where to paint a pattern on a screen in front of my eyes that will align with the outside world.
My head spins just thinking about it, but the X marks the spot with some clever timed flashes
This parent shows how closely linked Veoneer and Seeing Machines must work especially for Human Factors.
My car likes to remind me to wear my seat belt, sometimes if I stay below 5 mph and in first gear I may get away with it for a while before a relaxing melodic chime alerts me, if I ignore it, 5 or 10 seconds later a new and more insistent melodic chime comes in louder. This is Veoneer in my car. The patent adds more steps like asking the driver to perform certain tasks to ensure that they have the drivers attention which they measure using the DMS and button presses, gestures, etc.
I would like to see what AlexRoi144 thinks of this as he has often said that a decent handover system was one of the things missing from Level 2+ systems. OF course this system doesn't solve the problem of urgent takeovers for issues the vehicle did not forsee!
Good find JC
so the aircraft was in an manual mode, since they took control and aborted the landing but I guess they forgot to press the TOGA button (Take Off Go Around) button to tell the aircraft what to do.
Mode confusion is quite common - but even with Seeing Machines on board, it will need a lot of extra development by the aircraft manufacturers to realise when no-one is actually flying the plane
Apologies for the 'astronomical' error. Don't worry the lens will flip the image.
This must be for a future generation of Magna mirror, the exploded view ee have seen so far didn't have room to move the lens.
Moving the lens wasn't the USP, that was having the lens not centralised with the centre of the sensor which is a sensible claim as most of the time the centre of the main objects of attention are below the axis of the mirror
The world conspires to make eye tracking difficult, wether, it is sunlight or glasses. The move to 940nm fixed many of the problems of sunlight, but it didn't so drivers from wearing sunglasses.
Now most glasses and sunglasses are not a problem. But when it just has to work in automotive "most " is not enough. So this patent combines ir and rbg in one sensor and adds a switchable polarisation filter.
Now I once had to make a switchable lens and it wasn't fun or easy, so my first thought was they are crazy. But, this isn't hard they just need to switch the polarisation, that is easy, already mass produced, miniaturised and most importantly cheap!
You, may already have one on your wrist now, that is how LCD screens work but instead of a seven segment number or pixel matrix this just needs a single small panel with two electrodes. No moving parts and already standard technology made to Automotive standards.
The relatively poor contrast ratio isn't a problem here, the current draw is low and only required if the driver's glasses need it.
So extra cost if a few cents will help to catch a few percentage points of edge cases
This is a neat idea, sensors are relatively cheap but processing isn't!
So add in a 3d Time Of Flight camera as well as the OMS camera. When you use the ToF camera it takes 300 GigaFlops to build the 3d picture of the occupants. Once that is done, switch to the 2d camera and model the 3d pose of the occupants from an AI model on the 2d image. That only costs 50 GigaFlops. So rather than specify processing to be larger they only go 3d occasionally such as when the models no longer coincide. That leaves the processing available for other ADAS tasks.
What do you need it for?
-airbag deployment
- directing air flow
-3d sound or sound cancelling or individual media for different occupants etc
Good find JC
Flight Crew Fatigue And Controlled Rest Management System
It is official TLS has lost it, but I found it in my reading pile and it was worth the wait.
So for many, many years Seeing Machines have been watching eyes to keep people awake, so surely throwing DMS into an aircraft ****pit , can't be that hard?
Well, you know that Human Factors thing they are always banging on about, well they know a thing or two about humans, but this lets them show a whole new side - watching people sleep, telling them when to sleep to be most effective how long to sleep for and when the need to be woken, so that they are rested at the key stages of a flight and truly alert. But wait, there is more. When two pilots are in an aircraft and the cabin doors are closed and they are pushing back from the airbridge, taxing across the apron, launching into the sky and all the way to cruising altitude, they are not discussing what they did at the weekend, no, it is professional clipped phrases and counter checks as they handle this risky phase of the flight - you don't want to miss an alarm or flashing warning discussing what was on the barbie. Well seeing machines is going to cover that too, lip reading to see if there is a conversation and does it coordinate with what is happening in the ****pit!
Now on to Tablets, we may think of the iPads that Pilots carry are just toys, but now they replace the mandatory stack of checklists, manuals and logs, they can manage weather, routing and navigation - this is the Electronic Flight Bag (EFB) device. Now they can do far more! "the aircraft interfacing device may include an Intelisight™ Aircraft Interface Device (AID) developed by Collins Aerospace, which is capable of storing and accessing aircraft data and communications, navigation logging, flight tracking, obtaining weather information and monitoring aircraft health among other things."
So the Collins solution has the Pilots interfacing with the outputs from Seeing Machine's computation in their EFB, they can confirm the flight plans and depending on how the pilots are feeling and how tired they are judged to be, they can plan their breaks and even sleep in a horizontal bed. This information is used to plan which pilot will be landing the plane. Over successive flights, you can track where pilots were tired and how they performed - this can even be used to adjust rostering.
Incidentally, SEE won't be watching closed eyes in a separate horizontal bed crew rest area, that is down to a worn device - does this sound familiar? Quantas' Project Sunrise has already tested most of the equipment in real flight conditions and now that Collins can sell it along with their Intelisight™ Aircraft Interface Device.
I am not sure that all pilots will get a flat bed, most will have 20-40 minutes of snooze in their chair at the ****pit (However, I wonder if FedEx pilots will rearrange the parcels to make a bed?).
A350-1000 ****pit image?
Thanks JC - another great find and
Just because someone has the cash to buy 22m shares doesn't mean they have the Bot, the guaranteed backup Internet connection with trader and tech support on stand-by, so the purchasing is done by the professionals dropping the price by selling constantly, while they are hoovering up the shares. This pair of matched transfer records that aren't used for pricing is proof as that deal is completed. Since the Bot was not running using the account of the eventual buyer.
Will they be back tomorrow? Or next week? One thing we do know if that the new owner of 22 million shares doesn't want the price to stay down here unless they plan to but more again soon.
Wonder who does their DMS?
https://news.gm.com/newsroom.detail.html/Pages/news/us/en/2022/aug/0803-supercruise.html
Original source!
To date, GM customers have driven more than 34 million miles with Super Cruise To date, GM customers have driven more than 34 million miles with Super Cruise engaged
Thanks Numpti,
I have known about this principle for 30 years and it has been on my watch list as active technology for the last year or so.
But "working with BMW" can mean 10 years from production.
It will take some time and a lot of money to make this a reliable alternative to the full features we use today.
One key clue was resolution 128 x 128 is a huge drop from the mega pixel cameras we use now.
JC, as you will know Intel own Mobileye, who are not direct competitors. It was interesting that this patent didn't have any obviously Mobileye staff (they are based in Israel), so this real Intel.
As for the content, this is allowing users to add themselves to the DMS model, so that it can better detect their face and state without sharing your face with those nasty DMS providers.
The reason given was certain illnesses can make drivers have "sore looking eyes" so the DMS system complained that they were looking tired. It is no doubt a very technically valid security system to build custom models and to store these securely in the vehicle and the cloud and tie them only to the identity of the driver etc , ....
But, In the images of a DMS was a face that I recognised from Cipia, the image even said Eyesight (which was their previous name) so the inability of Cipia to realise that a driver with Sjogren Syndrome was tired because their model saw sore eyes as tiredness leads to a vastly overcomplicated technical solution that requires so much trust in the OEM to support it in 15-20 years as well as the vehicle and the necessary mobile apps - that was your cue to spot that Silicon Valley doesn't do long term support especially not on mobile!
The alternative to phoning home to Intel to fix Cipia's issue would be to try a better DMS provider that doesn't use blinking as a signal for tiredness.
I have posted in the past about Mobileye's more recent hardware being suitable for hosting Seeing Machines. I wonder if that made this patent as pointless as its fragile suggestion?
If Mr Robot tries to sell all his new shares (without buying them back) we will have a brown liquidity problem
Not suspicious at all, entirely random and coincidence that they occur in groups at similar times as though they are part of a random plan to push the price down.
Not at all related to the groups of purchases that are happening now that bought 300k at the new knock down price, move along now, nothing to see. Important people are making money. If you suspect something, you are just imagining it.
- if you have funds, take advantage of it
- if you don't have funds, it won't last forever as it is costing someone money
- if it is costing someone money, then at the end of their purchases, they will want to boost the price to where it should be or beyond
- they are in for the long haul, as they know they can't sell out at the peak without crashing the price (unless it is Elon and his army of minions buy in at the top)
Apologies for the delay, I was away so didn't see the bat signal on the Canberra horizon. I haven't had time to digest the whole patent - it isn't much fun reading it on a mobile, so just a few points for now.
When I was growing up, we thought by the new millenium, we would all be wearing shiny suits and be driven around in flying cars by our robots. But reality got in the way and now we need to babysit our robots which are only really capable of steering and and controlling the speed - assume they can work out where the road lines are and they correctly guess the behaviour of EVERYTHING on the road. Of course babysitting is boring, you have to stay awake, and you can't even have a drink or watch a film in the car - unless you cheat., since we are still the responsible adult in the room and we can either die or face legal repercussions for not supervising the driving properly, or doing it ourselves the old fashioned way we need help behaving
So how to Seeing Machines make sure the humans have skin in the game - well they look for our skin! You see, it is possible to fool a camera by showing it a picture, so how do we avoid needing 2 cameras (and 3-d processing) well we can modify the lighting. I will pick one "simple" option. Imagine a flat light source that magically contains vertical stripes like pinstripes. If you project this light onto a sheet of paper, you will still get straight lines. View the paper from an angle and it still has straight lines. Now imagine a face, the lines will work like contours as long as they are viewed at an angle from the illumination. Change the light source so that the lines are horizontal (assuming the camera is to the side) and the effect almost vanishes in line with the camera, but will still be visible as you move up or down.
OK, next step, use a grid of light spots. On a flat sheet of paper, the image of the spots might be 100 pixels apart. Now put a football in the light instead of a head. Viewed slightly to one side, we see the central square is probably still around 100 pixels, but as you move away, either the width or thickness of the squares becomes larger and the shapes are no longer square.
Now you can draw a graph these distances. On the flat sheet of paper, everything is around 100 pixels, so there is a huge spike. Change to a ball and the range of measurements varies greatly - but smoothly, and moving the ball or rotating it doesn't change the pattern much. OK now use a face and the complex "topology" gives a far wider spread of values. Flat foreheads, chins and brows may be mostly flat, but as the head turns, these will distort the squares and result in longer or shorter distances.
That is called a Fast Fourier Transform and is relatively easy to develop and process, but it can tell a flat picture from a curved picture, a ball or a head quite easily, if you can feed it a "spatial pattern".
I didn't get to "skin" or speckles yet, but I am out of characters, so that is a story fo
You could apply some poster's opinions to Tobii's approach to DMS. A successful company branching into an area that looks like a simple extension to grab a large market share.
Let's compare:
Aviation and Automotive are slow to move, highly regulated industries but there are openings if you have the correct use case and the partners to open the door.
Tobii hopes a few flyers and a booth at InCabin will see them swept off their feet, but where are the existing partners? Tier1s are already signed up and they don't have a path to market.
Now look at SEE in aviation, they have already done the hard miles, the tests and the validation. In the air and in simulators, not just as demoes, but as comperative products with costed benefits. The seeds are planted and when the aviation market improves they will grow and spread wide. Collins, Boeing, L3Harris, Airbus don't need to do trials with the likely lads, they already have a "Vetted" supplier who they know, trust and already work with