AcuityPercept leverages a global directive and local refinement algorithm, utilizing metadata and loss feedback from AI task models to achieve optimal ISP tuning automatically. By continuously refining ISP settings, it enhances object detection accuracy and delivers optimal ISP-processed images for neural network processing. This comprehensive optimization solution, combined with VeriSilicon’s ISP IP, enables automated closed-loop optimization.
Key Updates and Features:
Enhanced Object Detection: AcuityPercept improves object detection accuracy by up to 30% compared to traditional ISP tuning methods
Efficiency Gains: The system reduces the time required for ISP tuning by 50%, allowing faster deployment of AI vision solutions
Broad Applicability: AcuityPercept is applicable to a wide range of AI-powered vision applications, including autonomous driving, robotic vision, and AIoT
“AcuityPercept is a key ISP tuning system that bridges the ISP with the target AI perception engine to achieve optimal object recognition,” said Weijin Dai, Chief Strategy Officer, Executive Vice President, and General Manager of the IP Division at VeriSilicon. “As AI-driven perception becomes increasingly critical in autonomous vehicles, smart surveillance, and robotics, our AI-powered automatic ISP tuning system lays the foundation for more accurate, efficient, and scalable AI vision solutions. AcuityPercept has demonstrated excellent results with road data collected by our automotive ISP customers, enabling vision perception engines to clearly and reliably identify objects.
For more information. Visit https://www.verisilicon.com
