Edge AI Based 3D Perception
ABSTRACT
Robotics, automation, and intelligent machines require accurate 3D perception to operate safely in the physical world. However, today’s depth-sensing technologies—such as LiDAR, Time-of-Flight, and stereo vision—often require expensive hardware, significant compute, and high power consumption, limiting their scalability for edge AI systems.
MagikEye introduces a scalable architecture for real-time 3D perception using its patented Invertible Light™ Technology (ILT). By combining standard RGB/IR camera sensors with structured illumination, the system converts low-cost camera modules into high-performance depth sensors. Efficient algorithms reconstruct depth with minimal compute, enabling deployment on edge processors like micro controllers without specialized hardware.
Multiple edge-based sensor modules can be integrated and fused through an AI perception SoC to produce real-time spatial understanding of the environment. This enables applications such as spatial mapping, object detection, gesture recognition, and robotic interaction.
Compared with traditional depth technologies, the MagikEye approach significantly reduces system cost, compute requirements, and power consumption while delivering high frame rates and low latency suitable for real-time applications. This architecture enables scalable spatial intelligence across consumer devices, robotics, autonomous systems, and next-generation AI platforms.
BIOGRAPHY
Skanda Visvanathan leads business development at MagikEye, focused on scaling Invertible Light™ (ILT), an ultra-efficient 3D sensing technology for edge AI systems.
Skanda has over 20 years of experience in semiconductors and advanced sensing technologies.
He previously served as VP of Business Development at ams AG and as VP and General Manager of the 3D Imaging Solutions Business Unit at Heptagon, where he led the commercialization of 3D sensing technologies. Earlier in his career, he heldsimilar roles at Hewlett-Packard and Agilent Technologies.
He holds a Master’s degree in Communication Engineering from Hamburg University of Technology and an MBA from University of Warwick.