Ai Sensing Company Overview

AI Sensing will exhibit at the "Kyoto Smart City Expo 2025" held in the Keihanna Research and Development City from October 2 (Thursday) to 3 (Friday). At our booth, we will demonstrate the management of road traffic conditions captured in images using the AI vision system GS500 from AglaiaSense.

Leopard Imaging's custom-designed Holoscan sensor bridge module is a plug-and-play interface that can simultaneously connect up to five 10GigE cameras directly to the Holoscan sensor bridge developer kit. It provides deterministic and synchronized data streaming using hardware-level timestamps, ensuring tight integration between vision pipelines. This bridge maximizes the high-speed sensor performance of systems built on IGX Orin by enabling real-time parallel vision workloads for AI inference, visualization, and data logging. Leopard Imaging's Holoscan-based solutions are designed to meet the demanding real-time imaging needs of robotics, automation, and industrial inspection, enabling ultra-low latency streaming of multiple cameras to NVIDIA IGX, NVIDIA's sensor-over-Ethernet streaming platform for edge AI.

The Raspberry Pi 5 and Leopard Imaging camera have further expanded the possibilities of vision. The Raspberry Pi 5 has finally arrived, marking a significant advancement for manufacturers, engineers, and technologists. With a faster quad-core Arm Cortex-A76 CPU, upgraded MIPI CSI interface, and significantly improved GPU and memory bandwidth, it is an ideal platform for those building vision-based projects. And now, thanks to Leopard Imaging's new Raspberry Pi 5-compatible camera, it is set to get even better. The enhanced processing power of the Raspberry Pi 5 and its high-speed camera interface are perfect for high-resolution, fast, low-latency image processing. By combining this with our MIPI camera module, we can achieve platforms such as: - Real-time computer vision - Robotics and drone navigation - AI-based object detection/tracking - Automotive prototyping - Industrial and scientific image processing.