Grove Vision AI Module V2 for Seeed Studio XIAO

The Grove Vision AI Module V2 is an MCU-based vision AI module powered by the Himax WiseEye2 HX6538 processor, featuring a dual-core Arm Cortex-M55 and an integrated Arm Ethos-U55 neural network unit. It supports both TensorFlow and PyTorch frameworks, and is compatible with the Arduino IDE for no-code model deployment and immediate visualization of identification results via SenseCraft AI.

Features

• Powerful AI Processing Capabilities: Utilizes the WiseEye2 HX6538 processor with a dual-core Arm Cortex-M55 and integrated Arm Ethos-U55 neural network unit.
• Versatile AI Model Support: Easily deploy off-the-shelf or custom AI models from SenseCraft AI, including MobileNet V1, V2, EfficientNet-Lite, and YOLO v5 & v8. TensorFlow and PyTorch frameworks are supported.
• Rich Peripheral Devices: Includes a PDM microphone, SD card slot, Type-C port, Grove interface, and other peripherals.
• High Compatibility: Works with the XIAO series, Arduino, Raspberry Pi, and ESP32 dev boards, making further development straightforward.
• Fully Open Source: All code, design files, and schematics are available for modification and use.

How Does the Grove Vision AI V2 Perform?

Compared to Seeed's other MCU-based vision AI boards, the Grove Vision AI Module and XIAO ESP32S3 Sense, the Grove Vision AI V2 delivers a significant step up in processing power, making it a strong choice for high-performance vision processing tasks.

Looking for a complete kit with XIAO and camera options? Check out the Grove Vision AI V2 Kit. Want a ready-to-use compact AI camera instead? Choose the XIAO Vision AI Camera, or explore the reCamera Basic for a more powerful standalone AI camera.

Specifications

Applications

Earlier MCU-based vision sensors were limited by low computing power, typically only supporting simple tasks like face detection. The Grove Vision AI Module V2 goes beyond these limits, enabling functionality that was previously unachievable with traditional MCU-based vision AI sensors, including:

• Facial Keypoint Detection: Goes beyond basic facial recognition with detailed keypoint detection, accurately identifying features such as the eyes, nose, and mouth.
• Human Pose Estimation: Analyzes key joints and limbs to determine posture, identifying body parts like arms, legs, torso, and head. Useful for fitness tracking, posture correction, and motion recognition.
• Object Detection: Supports powerful models like YOLO v8n, capable of detecting and classifying over 80 types of objects.

How to Use It

SenseCraft AI is a platform for easy AI model training and deployment with a no-code/low-code approach. It natively supports Seeed products, ensuring full adaptability of trained models. Deploying models through the platform also provides immediate visualization of identification results on the website, allowing quick assessment of model performance.

Ideal for tinyML applications, adding vision AI to your smart sensor with the Grove Vision AI V2 is straightforward with SenseCraft AI. Connect the device, select a model, and view identification results with minimal setup.

Resources

• HX6538 Datasheet

This product can be found here:

• Accessories
• Accessories
• Arduino Shields
• Components & Parts
• Development Boards
• Hats & Other Add-ons
• Motors
• Power
• Seeed Studio
• Seeed Studio XIAO Expansion / Breakout Boards
• Servo Drivers
• Shields/Hats & Cases
• WiFi Components
• Wireless Connectivity