Arducam IMX500 Cameras are a versatile collection designed for both proof-of-concept and real-world deployment. Powered by Sony’s IMX500 sensor and AITRIOS™ platform, they combine edge AI intelligence with flexible interfaces—including PoE, USB, SPI, and GSML2—to adapt to diverse project needs. With options ranging from Raspberry Pi AI Cameras for seamless integration into the Pi ecosystem to industrial-grade designs built on Compute Module 5, Arducam delivers a one-stop solution to accelerate vision AI innovation across industries.


Supports flexible M12 lenses from 20° telephoto to 180° fisheye to perfectly match any field of view.
Under a 25°C ambient environment, after one hour of continuous AI inference, the camera module stabilizes at 34.8°C and the IMX500 at 35.0°C, corresponding to a temperature rise of approximately 10°C above ambient.
Fully compatible with official Raspberry Pi AI examples and frameworks with zero code changes required.

The first robust solution to bring full IMX500 AI performance to microcontrollers (ESP32-P4, Arduino, etc.)
Unique design featuring both SPI, I2C, MIPI and USB outputs, offering maximum compatibility across low-end MCUs and high-end processors.
On-sensor processing drastically reduces data transmission load, enabling ultra-low power consumption ideal for battery-operated IoT edge devices.

Integrated with various robust processor like Raspberry Pi CM5, CM4, and cost-effective CM0 version, so ready for deployment.
Supports CCM and M12 lenses, matching different optical needs.
Supports industry-must features like POE/USB, integrated fan cooler, and GPIO potential.

Integrated with various robust processor like Raspberry Pi CM5, CM4, and cost-effective CM0 version, so ready for deployment.
Deploy immediately, IP66 waterproof, dustproof and fogproof, PoE power supply.
Supports industry-must features like POE/USB, integrated fan cooler, and GPIO potential.
Why your Edge AI model might be underperforming (and it’s not the code).
You’ve optimized your neural network and selected the powerful Sony IMX500 sensor. But when you deploy it, the accuracy drops. The culprit? The lens.
In embedded vision, the golden rule is “Garbage In, Garbage Out.” If your optics introduce severe distortion or lack the required pixels-on-target, even the most sophisticated AI model cannot magically reconstruct the missing data

Arducam IMX500 M12-mount AI Camera supports a versatile range of industrial-grade M12 lenses to perfectly match your target distance and scenery
20° HFOV Telephoto: For long-distance detail magnification.
50° HFOV Low-Distortion: For precision metrology and scanning.
72° HFOV Standard: Balanced FOV for general machine vision.
88° HFOV Wide-Angle Low-Distortion: Expanded view without geometric warping.
140° HFOV Ultra-Wide: Larger Field of view for humanoid robots, etc.
180° HFOV Fisheye: Maximum situational awareness and panoramic tracking.
Build AI Vision prototypes in days, not weeks.
Affordable One-stop Vision AI Solution, from evaluation to deployment.
Trusted by and collaborated with Sony for End-to-End smart solutions.
From advanced sensor tuning to mass manufacturing, we support every stage of embedded vision development —all under one roof.
Our services have been widely trusted by engineers, researchers, system integrators, and OEMs across industries such as robotics, industrial automation, smart agriculture, medical devices, and more.
Arducam is proud to collaborate with the most influential players in embedded computing and vision. From silicon vendors to platform providers, our partnerships help ensure that our camera solutions are well-aligned with the latest technologies and workflows.
Gaze Detection, People Flow Analysis, Shelf Monitoring
Quality Inspection, Anomaly Detection
Warehouse AMRs, OCR, Inventory management
Mechanical Arms and other robots that require long distance between cameras and platforms.
Whether you want to detect objects, recognize images, estimate poses, or segment scenes, this collection offers ready-made AI models and demo examples designed for a variety of real-world applications. Explore the Raspberry Pi Model Zoo and the Arducam Model Zoo to find, test, and deploy the AI models that best fit your project needs — no deep AI expertise required. It’s all about helping you quickly bring smart vision capabilities to your device.

When off-the-shelf isn’t enough, Arducam offers end-to-end custom camera design services to help turn your ideas into real products. From early concept to mass production, we work closely with you to tailor imaging solutions that meet your unique technical requirements and product goals.