OmniVision Technologies, Inc., a leading developer of advanced digital imaging solutions, Ambarella, Inc. (Nasdaq: AMBA), an AI vision silicon company, and Smart Eye, a world leader in developing AI-powered eye, mouth and head tracking technology, today announced in advance of AutoSens Detroit the automotive industry’s first complete solution for dual-mode camera applications. This joint solution simultaneously monitors drivers while capturing vehicle occupants for one-way videoconferencing. It features OmniVision’s automotive industry-first OV2312 image sensor, with a dual-mode global shutter that captures both RGB color images and IR images under low-light conditions. These dual captures are then processed simultaneously by Ambarella’s CV22AQ CVflow® computer vision processor, which runs Smart Eye’s algorithms to analyze the driver’s state and alert the vehicle to any unsafe indicators, such as drowsiness. The companies worked together to integrate and fine tune this solution, which is available as a complete reference design, allowing automotive designers to focus on differentiating their final application while simplifying the overall design effort.
The companies are currently in discussions with automotive OEMs regarding the implementation of this reference design in upcoming car models. This joint solution enables the full range of driver and cabin monitoring implementations, from safety applications to augmented reality displays and interior sensing with deep neural networks and AI. The solution’s one-way video conferencing allows remote conference participants to see those in the vehicle, while the vehicle receives only audio to reduce driver distraction.
“Our CV22AQ offers best-in-class image processing and high performance AI computing at low power consumption, typically below 2.5 watts,” said Fermi Wang, president and CEO of Ambarella. “While we have worked together to pretune this joint solution, designers still have the flexibility to further customize it to the needs of their specific implementations.”
“Our AI software makes it possible for automotive OEMs and tier-1s to deploy a new generation of driver and in-cabin monitoring systems with advanced AI features, increasing safety and convenience for drivers and passengers,” said Martin Krantz, CEO of Smart Eye. “This dual-mode joint solution has the unique ability to monitor the driver while simultaneously capturing images for videoconferencing, regardless of lighting conditions.”
“Our OV2312 is the only automotive image sensor that offers the combination of dual-mode RGB-IR capture and a global shutter,” said Mario Heid, vice president of OmniVision Europe. “Without this combined functionality, the camera system designers would have to use two separate image sensors, which is too expensive for most vehicles. At the same time, the European Union (through Euro NCAP) is requiring that all new cars sold in Europe have a driver monitoring system (DMS) camera by 2022, and our joint solution gives OEMs the flexibility to differentiate their vehicle features while meeting this mandate.”
The OmniVision OV2312 is a dual-mode automotive image sensor, enabling single-camera driver state monitoring and viewing applications, such as the videoconferencing capability in this reference design. It remains the automotive industry’s only RGB-IR global shutter image sensor, while offering the smallest size in its class of 2MP sensors. The OV2312 provides advanced ASIL functional safety and industry-leading near-infrared (NIR) light performance of 14% quantum efficiency at the 940nm wavelength, along with excellent modulation transfer function (MTF). Its power consumption is also industry leading, consuming just 190mW in typical conditions. The end result is that the OV2312 reduces the number of image sensors in the system, saving both power and space by eliminating the need for multiple rolling shutter sensors.
The OV2312 is the latest member of OmniVision’s automotive global shutter image sensor family. It is pin-to-pin compatible with the OV2310 and OV2311 family members, which provides designers with the flexibility to choose the best feature set for their needs.
Ambarella’s scalable CVflow processor family provides a full range of systems on chip (SoCs) across multiple performance and price points. The company’s powerful image signal pipeline (ISP), with support for OmniVision’s dual-mode RGB-IR color filter arrays, enables high-accuracy detection and monitoring, even in low-light, in-cabin environments. Its high dynamic range (HDR) processing extracts maximum image detail in high-contrast scenes, further enhancing the chip’s computer vision capabilities and the performance potential of Smart Eye algorithms.
Smart Eye’s DMS solution offers eye tracking software for integration in passenger cars and other vehicles to facilitate better safety and other functions that improve the user experience. By carefully studying eye, facial and head movement, Smart Eye’s interior vehicle algorithms can draw conclusions about a person’s alertness, attention and focus.
This joint reference design, including a demo board equipped with the three companies’ pretuned devices and software, is expected to be available to qualified customers later this quarter. Customers interested in obtaining this 3-way demo should contact their local sales representative for one of the three companies. For more information, contact your OmniVision sales representative: www.ovt.com/contact-sales.
A live virtual demo and interactive Q&A session for this joint reference design will be available at AutoSens Detroit Online, on Wednesday, November 18th at 2:50pm (Eastern). Event information and registration are available here: https://auto-sens.com/autosens-detroit-edition/.
Ambarella’s products are used in a wide variety of human and computer vision applications, including video security, advanced driver assistance systems (ADAS), electronic mirror, drive recorder, driver/cabin monitoring, autonomous driving and robotic applications. Ambarella’s low-power system-on-chips (SoCs) offer high-resolution video compression, advanced image processing, and powerful deep neural network processing to enable intelligent cameras to extract valuable data from high-resolution video streams. For more information, please visit www.ambarella.com.
About Smart Eye
Bridging the gap between man and machine since 1999, Smart Eye develops artificial-intelligence (AI) powered eye tracking technology that understands, assists and predicts human intentions and actions. By studying a person’s eye, face and head movements, Smart Eye technology can draw conclusions about an individual’s alertness, attention and focus, as well as gain insights into a person’s awareness and mental status. Today, Smart Eye’s eye tracking technology is embedded in the next generation of vehicles, helping the automotive industry take another step toward safer and more eco-friendly transportation. Smart Eye research instruments offer unparalleled performance in complex, real-world situations, paving the way for new insights in aerospace, aviation, psychology, neuroscience, medical and clinical research. Smart Eye is headquartered in Gothenburg, Sweden, and has offices in Michigan, USA, Tokyo, Japan and Chongqing, China, as well as having partners, resellers and distributors in Europe, USA and APAC. Its solutions are used by more than 700 clients all over the world, including leading research groups, brands and labs such as the U.S. Air Force, NASA, BMW, Lockheed Martin, Audi, Boeing, Volvo, GM and many more. For more information, please visit http://smarteye.ai.
OmniVision Technologies, Inc. is a leading developer of advanced digital imaging solutions. Its award-winning CMOS imaging technology enables superior image quality in many of today’s consumer and commercial applications, including mobile phones; security and surveillance; automotive; tablets, notebooks, webcams and entertainment devices; medical; and AR, VR, drones and robotics imaging systems. Find out more at www.ovt.com.
OmniVision®, and the OmniVision logo are trademarks or registered trademarks of OmniVision Technologies, Inc. All other trademarks are the property of their respective owners.