On 1 December 2024, both
@Humble Genius and
@Kamikaze kai shared the following LinkedIn post by Sounak Dey from TCS Research with us:
Sometimes back, our work focus was on the use case of "In-car Driver Drowsiness Detection". We did the usual tasks of designing a Spiking Neural Network (SNN) and then deploying it on BrainChip Akida 1000 neuromorphic board. The model has achieved upto 92% accuracy on four classes. Energy...
www.linkedin.com
View attachment 83536
View attachment 83540
Yesterday, the work he referred to, which was presented at ICPR 2024 in Kolkata (1 - 5 December 2024) by Sounak Dey’s and Chetan Kadway’s co-author Ajoy Dey, was published online (the text beyond the abstract is, however, behind a paywall):
Car accidents due to driver drowsiness is a very serious pan-world problem. At present there are few AI-based drowsiness detection and alert systems, but they are not suitable for cars due to their requirement of large memory, power, and latency and...
link.springer.com
View attachment 83537
View attachment 83538
View attachment 83539
Speaking of driver fatigue and distraction detection:
A so far underrated player in the Edge AI space is FotoNation.
This computer vision AI solutions company headquartered in Galway, Ireland, with a second engineering office in Brasov, Romania, was originally founded in 1997. In 2008, it was acquired by Xperi (then known as Tessera), and subsequently in January 2024 by Tobii, which calls itself “the global leader in eye tracking and pioneer of attention computing.”
Only ten months later, though, in November 2024, FotoNation was re-launched as its own brand after “a Management Buy Out led by three of its legacy executives Petronel Bigioi, Sumat Mehra and Eran Steinberg.”
These industry veterans behind FotoNation have now firmly set their eyes on the Edge AI market, offering “ultra-low power embedded Edge AI solutions”. Given they have successfully commercialised their tech in the past, they shouldn’t be dismissed as just another startup that has to build up an ecosystem from scratch.
Login to LinkedIn to keep in touch with people you know, share ideas, and build your career.
www.linkedin.com
www.fotonation.com
Imaging technology leader FotoNation has done now for the third time what other companies can’t do once: Complete a management buyout by the founders. FotoNation is a leading provider of ultralow…
thedeadpixelssociety.com
“Tobii has divested certain research and development activities, including assets related to chip-level hardware, firmware, and imaging technology, to the former founders of FotoNation. As consideration for the divestment,
Tobii holds a minority stake of 19.9 percent in the new company.
The divestment will result in a total reduction of around 35 employees.”
Yesterday, FotoNation posted on LinkedIn about their
vision to revolutionise road safety with the help of an AI-powered sensor in collaboration with their joint venture partner CameraMatics, having secured more than € 6 million in funding from Enterprise Ireland’s Disruptive Technologies Innovation Fund.
'The future of this technology, whether it is us or somebody else designing it, will be in every car in the world in about 10 years time,' says Tom Farrell of CameraMatics
www.irishtimes.com
Business
Car crashes due to fatigue could be ‘entirely eliminated’ by AI-powered sensor
Technology under development by two irish companies could also detect drivers on phones or looking away from road
A joint venture, between fleet-safety technology company CameraMatics, and FotoNation, a computer-vision AI solutions firm, aims to eliminate the need for cameras in driver fatigue and distraction monitoring systems. Pictured, from left, are Gabriel Costache and Petronel Bigioi of FotoNation, and Daragh McDonnell, Paulo Zanni and Rob Eviston of CameraMatics
Hugh Dooley
Fri Apr 18 2025 - 14:14
A new
AI-powered sensor that could prevent
car crashes caused by driver fatigue, distraction and illness has received more than €6 million in funding.
Two Irish companies behind the project have secured the financing from the Disruptive Technologies Innovation Fund.
The joint venture, between fleet-safety technology company CameraMatics, and FotoNation, a computer-vision AI solutions firm, aims to eliminate the need for cameras in driver fatigue and distraction monitoring systems.
Fatigue is a factor in as much as 20 per cent of all road traffic collisions, according to the European Road Safety Observatory, with mobile phone use a cause of as much as 30 per cent of all cases. The in-development technology is aiming to eliminate those crashes.
“What this technology does is it scans the cabin of the vehicle, analysing the driver in terms of their face and body to identify when someone is falling asleep,” says CameraMatics chief marketing officer Tom Farrell, with the product alerting or awakening the driver using sound.
The technology will also detect when someone is looking away from the road or is using a mobile phone or even having a medical issue, but without the use of cameras which, the company says, “can raise issues around privacy”.
Instead of cameras, the companies are hoping to use “neuromorphic sensing” – technology to create computer-intelligible images that are not readable by humans.
While CameraMatics operates similar fleet-safety technologies in commercial vehicles, it has identified the private car market as the focus for the new technology – in which, the company says, they have seen “immediate demand”.
“The future of this technology, whether it is us or somebody else designing it, will be in every car in the world in about 10 years time,” said Mr Farrell.
Mervyn O’Callaghan, chief executive of CameraMatics, said the development of the technology could “generate hundreds of tech jobs in Ireland and in other global markets CameraMatics operates in”.
Petronel Bigioi, chief executive of FotoNation, said the technology “could save thousands of lives around the globe” and has the “potential to entirely eliminate road traffic accidents caused by distracted driving and fatigue”.
Imelda Lambkin, Disruptive Technologies Innovation Fund manager at
Enterprise Ireland, said the technology “will dramatically impact road safety”.
The project is being developed in conjunction with the Center for Computational, Cognitive and Connected Imaging (C3I) at of Galway and the Centre for Advanced Photonics and Process Analysis (CAPPA) at Munster Technological University.
Nine months ago, researchers from FotoNation and the University of Galway published a paper titled “Spiking-DNN: Neuromorphic Event Camera based Driver Detection with Spiking Neural Network”, which proves they have been exploring SNNs:
While the above reference to future work with regards to evaluating their approach on neuromorphic hardware “such as the Loihi-2 chip from Intel” doesn’t exclude the possibility of it (also) being evaluated on Akida, we should consider them competition as long as not confirmed otherwise (eg by a partnership announcement).