Gravity HuskyLens 2 AI Image Processing Sensor AI Vision

HUSKYLENS 2 is a next-generation AI vision sensor designed to put powerful machine vision into the hands of every creator. Running on a 1.6GHz dual-core processor with an integrated 6 TOPS AI accelerator, it executes all algorithms directly on-device for high-speed, low-latency results. Its design focuses on three core strengths: Rich AI Model Support with over 20 built-in, ready-to-use algorithms and the flexibility to deploy custom-trained models; Flexible LLM Interaction, capable of both calling large models and providing local, real-world perception to Large Language Models (LLMs) via the built-in MCP service; and Wide Compatibility, ensuring seamless connection with a vast range of main controllers. Equipped with a replaceable 2MP camera, a vibrant 2.4-inch IPS touchscreen, microphone, speaker, and indicator lights, HUSKYLENS 2 is the ultimate tool for educational robotics, interactive systems, and smart device prototyping.

Lightning-Fast On-Device AI with 6 TOPS Power

At the heart of HUSKYLENS 2 is a dual-core 1.6GHz processor featuring an integrated 6 TOPS AI acceleration unit. Combined with 1GB of RAM, this significant computing power allows HUSKYLENS 2 to run complex algorithms and neural networks directly on the device. This facilitates real-time processing for demanding tasks such as multi-object detection and complex pose estimation, providing high-speed and low-latency performance without relying on cloud computing.

Start Creating Instantly with Over 20 Built-in AI Models

HUSKYLENS 2 is designed for immediate productivity. It comes pre-loaded with a rich library of over 20 AI models covering a wide range of common computer vision tasks, from facial recognition to object tracking. This extensive collection allows users to start building projects right out of the box with minimal setup, significantly lowering the barriers to AI application development.

Empowering LLMs with Contextual Awareness via MCP Service

HUSKYLENS 2 is the first AI vision sensor to feature a built-in MCP service, fundamentally changing how Large Language Models (LLMs) perceive the physical world. While other cameras provide raw image "snapshots," HUSKYLENS 2 offers a structured and intelligent summary. It can inform an LLM that a specific predefined person is performing a certain action. Bu gives LLMs accurate and context-sensitive perception, enabling them to make smarter and more informed decisions.

Deployment of Custom Models Trained with YOLO

Beyond built-in models, users can train their own vision models using YOLO and run them on HUSKYLENS 2. HUSKYLENS 2 comes with a complete set of labeling, training, and conversion tools that make creating custom-trained models easy.

Flexible Model Combination for Diverse Applications

The HUSKYLENS 2 architecture supports the execution of multiple models in series or parallel. This capability allows for the creation of more complex and detailed applications. For example, combining hand keypoint detection with object tracking can create advanced touchless control systems where a hand gesture can trigger the tracking of a specific object. This multi-model approach enables layered logic and more intelligent interactive behaviors.

Real-Time Video Streaming for Monitoring and Debugging

HUSKYLENS 2 provides seamless data collection and image display through its real-time video streaming function. It supports both wired (USB-C) and wireless (via optional Wi-Fi module) image transmission. This feature is critical for applications requiring remote monitoring, collecting datasets for training new models, and debugging visual applications by presenting a live view of what the sensor sees.

Swappable Camera Module for Every Application

HUSKYLENS 2 is equipped with a modular camera module that allows it to easily adapt to a wide range of scenarios, from microscopic inspection to long-range monitoring. Its unique flexibility supports interchangeable lenses, including manual focus, microscope, and night vision modules.

Wide Compatibility with Main Controllers

With standard UART and I2C interfaces, HUSKYLENS 2 provides seamless connection with a vast range of microcontrollers and single-board computers, including Arduino, micro:bit, Raspberry Pi, ESP32, UNIHIKER K10, and UNIHIKER M10.

Master Every Feature with the Comprehensive Education Wiki

HUSKYLENS 2 is supported by an incredibly detailed education wiki covering every single feature. This resource offers step-by-step tutorials for all 20+ models, advanced parameter settings, and specific code and graphical programming examples for Arduino, Raspberry Pi, micro:bit, and UNIHIKER.

Setup for Beginners: Powering on and changing language. Functional Deep Dives: Detailed guides for face recognition, object tracking, line tracking, OCR, and all other functions. Advanced Control: Settings such as recognition thresholds, NMS thresholds, and custom names. Model Management: How to import and export trained models for backup or sharing.

Model Hub for Ready-to-Use Vertical Models (Coming Soon)

To further accelerate development, a Model Hub will be launched. This online platform will host a curated collection of official and user-generated models tailored for specific vertical markets such as agriculture, retail, and security monitoring. Users will be able to browse, download, and deploy these specialized models directly to their HUSKYLENS 2 devices, providing instant access to powerful and niche AI capabilities.