# M5Stack UnitV2 AI Camera

> Linux-based standalone AI camera with SigmaStar SSD202D dual-core Cortex-A7, GC4653 4MP sensor, 128MB RAM, WiFi, and USB-C. Runs Python and V-Training for edge AI.

## Quick Facts

- **Brand:** M5Stack
- **Board Type:** ai-camera
- **Price:** ~$75
- **Product Readiness:** consumer
- **CPU:** SigmaStar SSD202D
- **CPU Architecture:** ARM Cortex-A7
- **CPU Cores:** 2
- **RAM:** 128 MB
- **Flash:** 512 MB NAND
- **USB:** USB-C
- **Power:** 5V
- **Operating Voltage:** 3.3V
- **Antenna:** Built-in PCB antenna
- **SD Card:** microSD

## Connectivity

- **Wifi:** WiFi 4

## Use Cases

- Computer Vision
- edge-ai
- Object Detection
- Face Recognition
- Machine Learning

## Components

- **SigmaStar SSD202D** (SoC): SigmaStar dual-core ARM Cortex-A7 @ 1.2GHz SoC with hardware ISP, H.264/H.265 encoding, designed for smart camera applications — [Datasheet](https://openhardware.directory/r?to=https%3A%2F%2Flinux-chenxing.org%2Finfinity2%2FSSD202D%2F&ref=openhardware.directory&product=m5stack-unitv2-ai-camera&source=md-export)
- **GC4653** (Camera Sensor): GalaxyCore 4-megapixel (2560x1440) CMOS image sensor with high sensitivity for AI vision applications — [Datasheet](https://openhardware.directory/r?to=https%3A%2F%2Fwww.galaxycore.com.cn%2Fproduct%2FGC4653&ref=openhardware.directory&product=m5stack-unitv2-ai-camera&source=md-export)
- **128MB DDR3** (RAM): 128MB DDR3 RAM integrated for Linux OS and AI model inference — [Datasheet](https://openhardware.directory/r?to=https%3A%2F%2Flinux-chenxing.org%2Finfinity2%2FSSD202D%2F&ref=openhardware.directory&product=m5stack-unitv2-ai-camera&source=md-export)

## Protocols

- **WiFi**
- **USB**
- **I2C**
- **Grove**
- **UART**

## Description

## Overview

The M5Stack UnitV2 is a standalone AI camera unit powered by the SigmaStar SSD202D dual-core ARM Cortex-A7 processor running Linux. It features a GC4653 4MP image sensor and is designed for edge AI applications including object detection, face recognition, and image classification.

## Key Features

- **SigmaStar SSD202D** dual-core Cortex-A7 @ 1.2GHz
- **128MB DDR3 RAM** and **512MB NAND flash**
- **GC4653** 4MP camera sensor (2560x1440)
- Built-in WiFi (2.4GHz)
- USB-C for power and data
- Runs Linux with Python support
- V-Training for custom model training
- Supports TensorFlow Lite and ONNX models
- Grove (HY2.0-4P) connector for peripherals
- microSD card slot for storage expansion

## Use Cases

- Object detection and counting
- Face recognition and tracking
- Barcode/QR code scanning
- Industrial visual inspection
- Edge AI prototyping
- Machine vision applications

## Where to Buy

- [M5Stack Official](https://openhardware.directory/r?to=https%3A%2F%2Fshop.m5stack.com%2Fproducts%2Funitv2&ref=openhardware.directory&product=m5stack-unitv2-ai-camera&source=md-export) — $75.00
- [OpenELAB](https://openhardware.directory/r?to=https%3A%2F%2Fopenelab.io%2Fproducts%2Fm5stack-unitv2-ai-camera&ref=openhardware.directory&product=m5stack-unitv2-ai-camera&source=md-export) — $47.90

## Resources

- [product](https://openhardware.directory/r?to=https%3A%2F%2Fshop.m5stack.com%2Fproducts%2Funitv2&ref=openhardware.directory&product=m5stack-unitv2-ai-camera&source=md-export)
- [github](https://openhardware.directory/r?to=https%3A%2F%2Fgithub.com%2Fm5stack%2FM5Stack_Linux_Libs&ref=openhardware.directory&product=m5stack-unitv2-ai-camera&source=md-export)

## Tags

`ai-camera`, `machine-learning`, `edge-ai`, `linux`, `computer-vision`, `cortex-a7`, `m5stack`, `object-detection`

## Images

![M5Stack UnitV2 AI Camera](https://nbg1.your-objectstorage.com/openhardware-directory/entities/m5stack-unitv2-ai-camera/40ad67ea29b8.webp)

---
[View full device page](https://openhardware.directory/devices/m5stack-unitv2-ai-camera)