M5Stack OV7740 UnitV K210 Edge Computing AI Camera

Resources:

K210 Datasheet
OV7740 Datasheet
Documents

K210 Edge AI Vision Camera with OV7740 Sensor, 0.8 TOPS KPU, and MicroPython Support
The M5Stack U078-C UnitV-OV7740 is a compact edge-AI vision camera built around the Kendryte K210 system-on-chip. The K210 integrates a dual-core 64-bit RISC-V processor running at 400 MHz, 8MiB of on-chip SRAM, and a dedicated neural-network processor delivering up to 0.8 TOPS for convolutional neural-network workloads. This enables image recognition and machine-vision processing to run locally without relying on continuous cloud connectivity.
The onboard OV7740 0.3MP colour image sensor provides a 65 field of view and supports VGA, QVGA, CIF, and smaller image formats. UnitV also includes 16MB Flash, two programmable buttons, a WS2812 RGB status LED, USB Type-C, HY2.0-4P expansion, and microSD card support for storing models and application data. MicroPython-based MaixPy development, together with Arduino and M5Stack vision tools, makes the module suitable for object detection, classification, shape recognition, tracking, and embedded machine-vision projects.
Key Features

AI Processor: Kendryte K210 edge-computing SoC
CPU: Dual-core 64-bit RISC-V RV64IMAFDC processor
Clock Frequency: 400 MHz nominal operating frequency
Neural Processor: Integrated KPU for convolutional neural-network acceleration
AI Performance: Up to 0.8 TOPS
On-Chip Memory: 8MiB 64-bit SRAM
Flash Memory: 16MB
Camera Sensor: OV7740 colour image sensor
Camera Resolution: 0.3MP
Field of View: 65
Image Formats: Supports RAW RGB and YUV output
Image Sizes: Supports VGA, QVGA, CIF, and smaller formats
Floating-Point Processing: Integrated double-precision FPU
Programmable I/O: Integrated FPIOA programmable I/O array
Security Acceleration: Hardware AES and SHA256 acceleration
DMA: Integrated Direct Memory Access Controller
Status Indicator: 1 programmable WS2812 RGB LED
User Input: 2 programmable buttons
Storage Expansion: TF/microSD card slot
USB Interface: USB Type-C
Grove Expansion: HY2.0-4P interface supporting I/O, UART, and I2C functions
Development Support: MicroPython/MaixPy, Arduino, V-Function, and V-Training

Applications

Object detection
Image classification
Real-time target localisation
Target size and coordinate acquisition
Shape recognition
Object tracking
Machine-vision systems
Embedded AI cameras
Robotics vision
Video recording
Industrial visual recognition
Educational AI projects
Edge-AI prototyping

Model Size

Package Includes:

1 x UnitV-OV7740
1 x HY2.0-4P Grove cable (20cm)