High Performance Scalable Sensor Hub DSP Architecture for Vision AI
Ceva-SensPro2 is the highly-scalable and enhanced 2nd generation high performance sensor hub DSP for multitasking...
Outstanding image quality and intelligent vision processing for embedded designs
From mobile devices to autonomous drones, and from surveillance equipment to automotive safety systems, intelligent image processing and computer vision are fundamental challenges for today’s system designers.
Demand for energy-efficient, compact, embedded imaging solutions is set to soar because low-cost, low-power computational photography and computer vision are essential to the success of the Internet of Things (IoT), where cameras and sensors reign supreme.
Efficient image processing is enabling this revolution. Meeting the advanced processing requirements of new and emerging vision requires advanced, flexible and scalable imaging and computer vision processing IP that:
- Provides a comprehensive solution for computational photography, artificial intelligence and computer vision applications, delivering a human-like visual perception
- Consists of vision and AI processors, deep neural network toolkits, real-time software libraries, hardware accelerators, and advanced algorithms as well as a strong algorithmic ecosystem
- Optimizes performance while reducing power consumption and die size, which enables efficient solutions for mass-market embedded devices
- Provides a low-risk, cost-effective development route that minimizes time-to-market
Target Applications
Mobile
Support for multi-camera, better low-light photography and intelligent features require dedicated AI and vision processors and advance algorithms
Automotive
Vision-processing IP plays a vital role in camera-enabled automotive safety, especially in autonomous vehicles and ADAS applications.
Surveillance
Smart cameras in surveillance and home security systems rely on advanced vision IP for video analytics and always-on operation done on the edge device.
Drones
Must-have features such as Sense, Avoid, and Navigate require sophisticated computer-vision capabilities, while keeping low power for better flight time.
AR/VR
Combining wide angle optics, efficient processors, advanced algorithms and sensor fusion enable high quality and law latency for best user experiences.
Wearables
Wearables devices demands optimized HW and SW imaging solution, operating with the lowest power and with the highest performance.
Sport & Action
Combining efficient AI processors, wide angle cameras and video analytics algorithms enables highest quality, advanced feature cameras with longer recording time.
Photography Gear
DSLR and mirrorless cameras demand state-of-the-art image and video quality, including better focus accuracy and image enhancement.
Radar Sensors
Advanced Radar front end processing platform, for both Automotive and consumer applications, cutting edge Radar AI and post processing
CEVA's industry-leading vision processing IP provides us with a comprehensive solution that enables us to integrate a host of innovative and intelligent system features into our ADAS product offerings.
Ross Jatou, Vice President and General Manager of the Automotive Solutions Division of ON Semiconductor