PR Center

News Center The Eyes of AI, Vision AI SoC 2023-02-22 1,220 hits

The Eyes of AI, Vision AI SoC
3 Core Technologies that VisioNexT focus on



The use of Artificial Intelligence(AI) is continuing to grow in various products and services. Across a wide range of AI technologies, the most prominent application field is computer vision, which works much the same as human vision. If AI enables computers to think, computer vision enables them to see, observe and understand. This is why vision AI system-on-chip(SoC) is necessary; in order to accurately acquire, identify, and process visual data such as digital images and videos. Furthermore, vision AI SoC plays an important role in machine vision systems that drive industrial automation.

VisioNext focuses on three main technologies to maximize the efficiency and performance of Vision AI SoC:






Powerful ISP

We generate clear video with the best image quality under difficult and complex environments. Built-in high-performance image processing technologies, including Low-Light, over 4K resolution, Wide Dynamic Range (WDR), Multi-channel camera support, Dewarping, and AI-based ISP.






Vision AI

To accelerate image processing technology, we design heterogeneous computing architectures with integrated Central Processing Unit(CPU), Graphics Processing Unit(GPU), Neural Processing Unit(NPU), and memory system configuration. This enables advanced AI-based object-detection, analysis, and processing performance.






Low Power

We minimize power consumption by designing SoC optimized for edge devices. Maximize power efficiency of IoT/embedded devices by using only the necessary processors (considering the condition), and leveraging DVFS (Dynamic Voltage Frequency Scaling).


Vision AI SoC will be playing a larger role in the AI solution from smart homes to logistics robots for automated warehousing. VisioNext will further develop the core advanced technologies to strengthen the growth engine of customer business.




List