Raspberry Pi AI Camera – 12.3MP Sony IMX500 Intelligent Vision Sensor

The Raspberry Pi AI Camera is a next-generation vision module built for on-device artificial intelligence, powered by the 12.3-megapixel Sony IMX500 Intelligent Vision Sensor. Unlike traditional camera modules, this AI Camera performs AI inference directly on the sensor, removing the need for external AI accelerators and significantly reducing system latency, power consumption, and CPU load on the host Raspberry Pi.

By capturing images and video streams enriched with tensor metadata, the AI Camera allows your Raspberry Pi to focus on higher-level application logic while the sensor handles vision inference tasks such as object detection, classification, tracking, and scene understanding. This architecture makes it ideal for real-time AI vision applications where efficiency and responsiveness are critical.

Fully supported by the Raspberry Pi software ecosystem—including libcamera, Picamera2, and the rpicam-apps suite—the AI Camera is accessible to beginners while remaining powerful enough for advanced developers. It maintains the familiar Camera Module 3 form factor, redesigned internally to accommodate the larger IMX500 sensor and optical subassembly, ensuring broad compatibility across all Raspberry Pi computers.

With features such as an integrated IR-cut filter, manual adjustable focus, wide field of view optics, and support for both RAW image output and AI tensor data, the Raspberry Pi AI Camera bridges the gap between traditional imaging and embedded AI—unlocking new possibilities in smart surveillance, robotics, edge AI, industrial monitoring, and research projects.

Key Features & Highlights

• 12.3MP Sony IMX500 Intelligent Vision Sensor
• On-sensor AI inference (no external accelerator required)
• Outputs images, video, ROI data, and tensor metadata
• Fully compatible with libcamera, Picamera2, rpicam-apps
• Camera Module 3–style form factor
• Integrated IR-cut filter and manual focus control
• Optimized for real-time edge AI vision applications

Technical Specifications

• Sensor: Sony IMX500
• Resolution: 12.3 MP (4056 × 3040
• Sensor size: 1/2.3" (7.857 mm)
• Pixel size: 1.55 μm × 1.55 μm
• Lens: 4.74 mm, F1.79
• Field of View: 66° (H) / 52.3° (V)
• Focus range: 20 cm – ∞ (manual)
• Input tensor size: Up to 640 × 640
• Data types: int8 / uint8
• Memory: 8,388,480 bytes (firmware + models + runtime)
• Frame rates:
  • 2028 × 1520 (2×2 binned) @ 30 fps
  • 4056 × 3040 @ 10 fps
• Output formats: RAW10, YUV/RGB, ROI, metadata
• Dimensions: 25 × 24 × 11.9 mm
• Operating temperature: 0 °C to 50 °C

Documents

• Raspberry Pi AI Camera product brief
• Getting started with Raspberry Pi AI Camera