Member Info for TheLongestShot

SYSTEMS AND METHODS FOR ENHANCED DISTANCE ESTIMATION BY A MONO-CAMERA USING RADAR AND MOTION DATA, not related to us, the term 'pose' and 'calibration' are common words in the field and do not need to stand on our patents in that area. Our patent was for using the tracking of human eyes to link the inside and outside worlds, which would calibrate the cameras. GMs patent are for external cameras and radars where the distance from the radar signal is used to calibrate the vision system's estimates of dimensions

In SYSTEMS AND METHODS FOR OBJECT LOCATION, the references to eye tracking etc is just one of many coverall statements trying to get this patent as wide as possible. No direct link to us, but like many of these patents that are primarily about vision systems, AV or ADAS, they will be using similar concepts, terms and ideas

JC, from a quick scan. Qualcomm #4 is just about the objects that it identifies. so that is can better and more efficiently classify and track them, including their potential directions of travel. Interesting, but not for us directly. however, it is relevant for the previous Qualcomm patent.

Klick,

It is good that Seeing Machines managed to provide the header image, but looks a like a city bridge with no trees in sight, and that truck wouldn't stay that clean if it had been in a forest. Maybe we need to "branch" out more with our stock images.
I know it isn't easy or cheap to plan ahead for all the future images you might need, especially now

Hi JC,

Just made it through the first Qualcomm patent for early fusing. It would certainly be possible for "someone skilled in the art" to take the SEE's gaze direction data, statistically collated to determine a frequency distribution of gaze angles. Then map those peaks in the frequency distribution and associate with corresponding gaze positions in the vehicle frame of reference.
It is only a short step to translate those locations to identify if these are the identified objects in Qualcomm's fused space. However, I didn't see any direct link between the patents, but they could certainly be used in one system to verify that the driver had observed the objects to enable a safe handover, or so that the car can "know" that the driver is aware of the other vehicle movements to prevent early nagging if there is an increased risk.

I might need a lie down in a dark room before I look at no #4

50 days free access to our @seeingmachines latest paper - on-road research w/ L2 automated vehicle??

Ft.??trust & acceptance; ??engagement; mode confusion; & strategies drivers use to workaround hands-on-wheel sensors?? https://t.co/LUf77Fpwsi

#humanfactors #driversafety https://t.co/JxMRWf9cKj

Safety Science
Volume 130, October 2020, 104845
Driver trust & mode confusion in an on-road study of level-2 automated vehicle technology

Kyle M.Wilson, Shiyan Yang, Trey Roady, Jonny Kuo, Michael G Lenné

Seeing Machines Ltd, 80 Mildura St, Fyshwick, ACT, Australia

University of Huddersfield, West Yorkshire, United Kingdom
Received 29 January 2020, Revised 11 May 2020, Accepted 19 May 2020, Available online 6 June 2020.

(finally beat JC to a patent ;)

That Denso patent references Gentex's patent for a rear view assembly with a camera facing the cabin for DMS:
http://www.freepatentsonline.com/y2019/0168669.html
VEHICLE IMAGING SYSTEM
United States Patent Application 20190168669Kind Code:A1

Abstract:
A system for a vehicle may comprise a rearview assembly comprising a display element configured for providing a rearward view of a scene behind the vehicle, at least one imaging device in communication with the display element and configured to capture images from within a passenger compartment of the vehicle, and a printed circuit board in communication with the display element and the at least one imaging device. The display element may be configured to selectively display content received from a source external to the vehicle. The display element may be suitable for receiving inputs from a plurality of sources. At least one of the plurality of sources may be one of the internet and a cellular network.
...
n some embodiments, dual-purpose imaging system 200 may operate in either a driver monitoring mode during which dual-purpose imaging system 200 monitors the driver of the vehicle, and a video-calling mode in which dual-purpose imaging system 200 captures images of a driver or passenger for video- or conference-calling. Dual-purpose imaging system 200 may be configured to alternate between the driver monitoring mode and the video-calling mode based on inputs received by dual-purpose imaging system 200.
...
In some embodiments, dual-purpose imaging system 200 may further comprise a controller 220 in communication with processor 214. In some embodiments, controller 220 may be in communication with at least one of light sensor 222 and image sensor 212, and may be configured to activate additional light source 218 upon receipt of certain inputs from at least one of light sensor 222 and image sensor 212. The certain inputs may include indications that there is insufficient ambient light for collecting images for video calling. Upon activation, additional light source 218 may project additional light onto the user.

When dual-purpose imaging system 200 is operating in a driver monitoring mode, it may capture images of the driver of the vehicle either intermittently or continually. Dual-purpose imaging system 200 may monitor and capture images of at least one of predefined facial, head, and body characteristics or positions of the driver. In some embodiments, the captured images may include but are not limited to: the position and movements of the pupils of the driver's eye(s), frequency and duration of blinking, position of eyelids, head position and orientation, head movement, pulse rate, facial and eye movements, or other suitable driver performance indicators.

In some embodiments, processor 214 may process the captured images to generate image data on driver characteristics. The generated image data may include, for example, blink rate, blink durati

United States Patent 10666901

Abstract:
A system in a vehicle comprises one or more sensors configured to obtain occupant information indicating a stress level of an occupant of the vehicle. The system also includes a wireless transceiver in communication with a mobile device including a camera configured to monitor the occupant, and a controller in communication with the one or more sensors and the wireless transceiver, wherein the controller is configured to determine a stress-level of the occupant utilizing at least the occupant information, initiate a video conference session on the mobile device between the occupant and one or more contacts of the occupant in response to the determination of the stress level of the occupant exceeding a threshold, and output on a vehicle display the video conference between the occupant and the one or more contacts.
...
An in-vehicle camera 103 may be mounted in the vehicle to monitor occupants (e.g. a driver or passenger) within the vehicle cabin. The in-vehicle camera 103 may be part of an occupant status monitoring system (OSM). The in-vehicle camera 103 may be utilized to capture images of an occupant in the vehicle. The in-vehicle camera 103 may obtain facial information about an occupant, such as eye-movement of the occupant and head-movement of the occupant, as discussed further below. The in-vehicle camera may be a color camera, infrared camera, or time of flight camera. The in-vehicle camera 103 may be mounted on a head rest, in the headliner, or located on a mobile device (e.g. tablet or mobile phone) to capture the driver's face, especially the driver's eyes.

A controller may receive driver status data from the OSM to determine an abnormal situation within the vehicle. The OSM employs one or more activity sensors such as a driver-facing camera, an occupant-facing camera, a health scanner, and an instrument panel to monitor activities performed by the driver or occupants (e.g. passengers in the vehicle). Based on the activity sensors, the OSM may determine whether the driver is, for example, distracted, sick, or drowsy as the abnormal situation.

A controller may receive occupant status data from the OSM to determine an abnormal situation within the vehicle. The OSM may employ one or more activity sensors such as a occupant-facing camera, a health scanner, and an instrument panel to monitor activities performed by occupants. For example, an occupant-facing camera may be employed in a vehicle headliner, vehicle headrest, or other area of the vehicle to monitor activity of the occupant. The OSM may also employ a mic that is in communication with a voice recognition (VR) engine that can detect fussiness of a child (e.g. baby) or irregular crying or sounds from the child. Based on the activity sensors, the OSM may determine whether the occupant or driver is, for example, fuss, experiencing motion sickness, hunger and fever.

http://www.freepatentsonline.com/10666901.html?TrucksFoT

(Fina

We will be a billion (Aussie) dollar company first. Fulfilling the prophecy.

Thanks JC, you never know when this may be useful:

"However, conventional systems that implement a 3D depth camera may still be spoofed. For example, placing a finger in front of a printed face picture may spoof a conventional system"

Worth watching the NTSB video with Missy Cummings that Colin includes in the article.

https://youtu.be/TqGv8PfZ7MQ
She really gets it.

Sg,

Yes it does mention Volvo Australia, they manufacture Volvo trucks in Australia and what he is directly referring to is the Monash University study which used Ron Finemore's drivers and their Volvo trucks. Volvo (trucks) Australia was a partner. unfortunately volvo group no longer owns Volvo cars. I presume that Guardian a not yet a standard factory fit in a new aussie volvo truck, but one day...

Robbie,
Yes like Renault - Nissan?
Smart Eye can have them.

Ok, I had a chance finally to review the GM patent fully and JC's favourite SEE patent - SYSTEM AND METHOD FOR IDENTIFYING A CAMERA POSE OF A FORWARD FACING CAMERA IN A VEHICLE.

I jumped the gun a bit when I saw reference to the camera moving to improve visibility, and guessed that it could be the visor or rear view mirror, but it was just the steering wheel/column that was moving along with the driverchair.

It just goes to show how powerful our IP is in this area. John Noble, Tim Edwards, Sebastian Rougeaux, James Henry, Grant Grubb and others have done a great job and I hope they have plenty of shares so that they can profit from their incredible work.

That's the one JC.

These are published in multiple countries with different references but I eventually found the matching number that was mentioned in the rejection of some of GM's claims.

When I get some time I will sit down on a proper screen and review them both. But it looks like GMs reason for the patent was the sunlight problem in the CT6. So they were looking to move to a movable part of the car (sun visor/rear view mirror?) that would need more calibration

This idea sound familiar, so I checked the Dossier:

"Claims 1, 4-5, 11, is/are rejected under 35 U. S. C. 102(a)(1) as being anticipated by Nobel et al (US201902064084)"

That name sounds familiar! JC will no doubt have a link.

The Qualcomm deal is the best and the worst for everyone in this board. Because it is licensing, we won't know about where it is being used and even if Seeing Machines did have an inkling they will be tied up in NDAs but unlike before when there would likely not be a 1-1 tie up with the eventual users and engineering fees, we would be working for Qualcomm. So we build a feature that could end up in a camera, a phone, a tablet, a Kinect, VR/AR headset/glasses, laptop.

Paul will be grinning like the Cheshire Cat who has had the cream and has the keys to the dairy, but unable to provide concrete facts so. We will still have clowns pouring cold water on the share price until the licensing revenue starts to flow and rise. But Paul won't be able to give firm forecasts. But, he will know the tsunami of money that will follow if the range of products grows.

New Microsoft x-box to launch around Nov 2020. Release date for VR not announced yet, no mention of next Kinnect yet.

https://www.techradar.com/amp/news/xbox-series-x

Whenever a great new tool or idea becomes available to science and technology it unleashes a new wave of science, creativity and technology that build on top. Whether it is the electron microscope, gene editing, Dewar flask or really good reliable eye tracking and driver state monitoring.

But while some technology can use an idea and move on without requiring to use the original stepping stones. You buy a phone but not the Dewar flask or electron microscope that were required for its development. But if you want to deliver on top of eye tracking, you still need to licence it or purchase it, and if you want to have the best you will be paying Seeing Machines.

S7 is at least putting a "Strong Smell" flag on his own posts