Can AI make cameras see in the dark?

Abstract

As cameras become ubiquitous in applications such as surveillance, mobile, drones, and automotive systems, achieving clear vision 24/7 under any condition—including extreme low light and high dynamic range scenarios—has become essential. By leveraging Edge AI processors, a software ISP based on neural network technology can process and optimize video in real-time, surpassing human vision capabilities in these challenging conditions.
In this webinar, we will explore how to harness the power of Ceva’s NPU alongside Visionary.ai’s advanced AI software Image Signal Processing (ISP) to significantly enhance video quality across various camera-enabled applications. Visionary.ai utilizes Neural Processing Units (NPUs) to boost camera performance, especially in demanding environments like extreme low-light and high dynamic range (HDR) scenarios.
Finally, we will introduce an available reference design for this joint solution and showcase the impressive results achieved.

Join Ceva and Visionary.ai experts to learn about:

• The power of Edge AI for computer vision
• Modern NPU architecture fundamentals
• True Night Vision software ISP
• The role of video denoising in solving the low-light challenge
• The technical challenges to implement an AI based ISP on an NPU
• How can Ceva and Visionary.ai enhance your camera enabled application

Target Audience

Vision system architects, hardware and software engineers, and product managers targeting camera enabled devices, that are looking to enhance their product ability to cope with extreme low-light conditions using the power of Edge AI.