Cipia, Ambarella leverage AI chip for a single camera in-cabin sensing solution
Cipia, an AI computer vision company focusing on automotive applications, has joined forces with Ambarella, an edge artificial intelligence solutions provider.
This collaboration aims to develop a monitoring system for vehicle manufacturers using only one camera within the cabin. The solution will provide automotive companies with a more cost-effective and flexible way to improve the safety of drivers and passengers, the companies say.
According to Cipia, its in-cabin sensing solution improves safety by incorporating a wide-field-of-view camera. By using Ambarella’s CVflow AI system-on-chip, this method enables the single-camera solution to monitor the entire in-cabin environment efficiently.
This means using advanced hardware SoCs, the in-cabin automotive system can process driver monitoring system (DMS) and occupant monitoring system (OMS) algorithms in real-time without compromising image quality or AI performance.
“We are proud to expand our collaboration with Ambarella, addressing together the in-cabin computer vision needs of automotive manufacturers by facilitating easy and seamless integration into spaces such as the rear-view mirror that require a small form factor and high thermal efficiency,” says Yehuda Holtzman, the CEO of Cipia.
Cipia’s DMS and OMS algorithms use edge-based computer vision and artificial intelligence techniques to establish a secure environment for drivers and occupants. The solution consists of two components: Cipia Driver Sense, which tracks the driver and their current state in real-time, and Cabin Sense, which monitors the overall in-car environment.
The solution utilizes Ambarella’s CV2 system-on-chip (SoC) family, which is optimized for processing video efficiently over a wide field of view (FOV). The device includes a 5MP RGB IR sensor that is specifically designed to work well in different lighting situations, making it suitable for in-cabin sensing solutions. By integrating Ambarella’s SoC and reducing the need for thermal management, this solution can be used in a variety of vehicle parts.
“The industry-leading efficiency and performance of our CVflow AI SoC architecture, combined with Cipia’s optimized DMS and OMS algorithms for WFOV cameras, provides automakers with a compelling option to add single-camera, all vehicle in-cabin monitoring within thermally constrained locations,” says Fermi Wang, the president and CEO of Ambarella.
Fast growing market for AI chips
Wang told Wall Street analysts on the company’s May 2023 earnings call that the CV2 family will be approximately 60% of total revenue in the fiscal year 2024. The line of chips will represent “a material portion of our operating profit dollars,” with an average selling price that is over double that of a typical video processing chip.
Ambarella recently announced the CV72S, a new system-on-chip (SoC) that allows security cameras to perform edge computing and utilize radar technology. This integration helps improve the accuracy of object detection, especially in low-light conditions.
Wang said the CV72S chips, a derivative of the next-generation CV3 AI chip design, will help the company expand its presence in the automotive industry while the market for Level 4 autonomous vehicles requiring the CV3 chip begin to come to market.
Article Topics
Ambarella | automotive | chip | Cipia | earnings | edge AI | sensor | SoC | video
Comments