The DSP has been deployed in Novatek’s newest generation NT98530 multi-sensor IP Camera SoC targeting surveillance, retail, smart city, and transportation. Novatek specialises in smart display and smart image total solutions, ranging from full line of display ICs and SoC for all types of display and image applications and, to date, has shipped more than 1 billion SoCs worldwide.
This latest engagement continues a successful partnership between the two companies, which spans almost a decade of licensing multiple generations of CEVA’s computer vision, audio/voice and AI DSPs.
“Today’s multi-sensor security cameras need powerful computer vision and edge AI processing capabilities to maximise performance and minimise the reliance on cloud connectivity for intelligent analysis,” said Novatek AVP Jimmy Su. “CEVA’s SP500 DSP delivers superior real-time performance for computer vision, AI-based analytics and multi-sensing sensor fusion at the edge, ensuring our customers have a powerful, flexible edge AI camera solution which they can customise to their requirements.”
Novatek’s NT98530 SoC is a highly integrated SoC with high image quality, low bitrate, and low power consumption. According to the company it provides the best cost/performance Edge-Computing IP camera solution currently in the market for multi-sensor security cameras.
The SoC supports up to 640M pixel/sec to achieve over 8MPixel/60FPS video performance while simultaneously performing advanced AI on each frame, with minimal power consumption. The integrated SensPro2, CEVA’s latest DSP generation along with its comprehensive SDKs is able to deliver up to 6X DSP performance improvement in computer vision, 2X AI inferencing improvement and 20% lower power compared to its predecessor, ensuring that Novatek’s latest SoC has the sufficient performance to support even complex computer vision algorithms and neural networks inferencing.
“We’re pleased to extend our collaboration with Novatek to enable their latest generation surveillance SoC support multiple cameras and the most advanced computer/vision AI use cases,” said Ran Snir, Vice President and General Manager of the Vision Business Unit at CEVA. “Our SensPro2 DSP is primed to support more of the camera analytics workload in real time at the edge, including the ability to perform vision/AI processing for multiple camera sensors in parallel, at minimal power consumption”.
