MAIX-III AXera-Pi

2022-12-08
Update history
Date Version Author Update content
2022-12-08 v0.1 wonder
  • Initialize Doc

If you think this AI development board are excellent enough and your friends may be interested with this, just share with them this link http://wiki.sipeed.com/en/m3axpi

maixiii_axpi

Official Online Store

In the past few years, many Linux AI chips come out.

Raspberry Pi and Jetson Nano these boards are really popular by their good performance, but now MAIX-III AXera-Pi comes out, equipped with the high-performance 3.6TOPs@INT8 chip ax620a, which will be your good choice and provide you a really good experience.

img

We design a core module for this chip, which is convenient if you want to design the ext-board for yourself.

MaixIII Core module
CPU 4 * Cortex-A7, with NEON and FPU
NPU 14.4Tops@int4，3.6Tops@int8
ISP 4K@30fps
Codec format H.264, H.265
Video encode 4K@30fps
Video decode 1080P@60fps
Ethernet Support RGMII / RMII Ethernet
Display MIPI DSI 4 Lane，up to 4K@30FPS
DRAM 2GB LPDDR4X 3733Mhz
Storge Optional 16GB EMMC (Boot from TF card by default)
IO pins DDR4 SODIMM 260P route all IO

Introduction

In WORLD ARTIFICIAL INTELLIGENCE CONFERENCE 2022, this AI board is shown the first time, like what's in the following picture.

And finally we launched this AI development board equipped with AX620A in 20221001. Its official name is MAIX-III AXera-Pi, while we like to call it M3AXPI in the following content.

So, why is this AI development board good?

  • Need documents? Follow our guides, to start journey from embedded Linux development to AI model deployment.
  • High power consumption? Just use one usb cable connecting with USB3.0 port to boot this device, no need of other power supply.
  • Difficult to develop? Compile your own application on this board, or train your own AI model online and deploy it on board.
  • Awful camera image? This chip is aimed for AI ISP night vision enhancement, we have adjust the camera to best state.
  • Too expensive? This board sold about 100$, cheaper than any other similar board.

Special column Documents

These documents are from Chinese club, read these via browser translator is a good idea.

Official Documents:

Community Documents:

Video reference:

Camera effect

Night vision enhancement

Left picture is original, right picture is enhanced

Exhibition darkroom

  • The left screen shows the enhanced picture and the righr screen is the original one.
exhibition_darkroom

Actual AI effect

AI Ecology

Support many Operators

many_operator

Screenshots of part operators. Click me to see whole operator support list

Many archived models

Before receiving this board, visit MaixHub to see your interested model, choose Platform axera-pi at the Lower left corner to find model can be deployed on AX-PI

maixhub_zoo.jpg

Apart from these, many models are uploaded to github AXERA-TECH/ax-samples, and we also put sme models in sustem image, just burn the system image into SD card and boot it on board to experience these models. Up to now the supported models are as follows:

Visit Deploy models to AX-Pi (Maix-III(M3) series) board to deploy models to AX-Pi.

Tranning your models

Tranning model online

Now MaixHub has been upgraded, with more friendly interface and more functions, here are steps to training model online.

Share your model

There is a model and codes or programs to load your model after you finish training, you can share these on Maixhub to others. Let's have fun and build the model Community together.

Cases

Updated in 20221205

Linux development

Run debian

We provide a linux image based on debian11, and you can boot this board from SD card after you burn the system image into SD card. We aim to make it easiler for user to use. Besides, it's really conveient to program AI development with Python API.

  1. Burning image: WIKI AXera-Pi Burn image

  2. Basic usage：WIKI AXera-Pi guide

  3. Program with Python：ax_pipeline_api

  4. Development Note：[maixpy3 axpi] How to port debian11 to ax620a (中文)

  5. Development Note：[maixpy3 axpi] Edit and pack the compiled debian image (中文)

BSP SDK

We provide source code, api documents, sdk usages, visit sdk usage to get more.

ax-sample

Here are typical AI models provided by Axera for development, evaluation and verification, those who are good at AI development or want to test these models can have a try, and these are not involved with peripherals.

  1. examples
  2. PP-YOLOv3
  3. PP-HumanSeg
  4. PP-Seg

libmaix

A unified embedded development environment come by SIPEED, includes camera, screen, vision, image processing and pipelines-related deployment examples, is suitable for those who want to learn embedded Linux.

  1. Test screen display
  2. Screen displays camera content
  3. camera + screen + yolov5 model

axpi_bsp_sdk

BSP development package for business, provided by Axera. It's the original development package for this chip, containing components like uboot、linux、msp 、sample、rtsp、ipcdemo, and this is progressively open source, you can get these codes for commercial evaluation, such as ipcdemo, but normally these codes are complex and highly coupled, those who are experienced enough and want to use this board for business cooperation can try this.

  1. Development documents docs
  2. Platform application ipcdmeo
  3. Peripheral Verificationsample

ax-pipeline

AX-Pipeline is provided by Axera. With M3AXPI, this project is used for people to know the usages of ISP, image processing, NPU, codec and display, which is convenient for developers to quickly evaluate and develop their own multimedia applications.

  1. Quick compilation Simple cross-platform compilation via cmake.
  2. How to replace into my own trained yolov5 model
  3. How to deploy my other models
  4. How to adjust the image orientation
  5. ModelZoo descriptions for models which are supported or to be supported

Hardware

Hardware function diagram

layout_axpi_2 layout_axpi_1

Core module parameters

MaixIII Core module
CPU 4 * Cortex-A7, with NEON and FPU
NPU 14.4Tops@int4，3.6Tops@int8
ISP 4K@30fps
Codec format H.264, H.265
Video encode 4K@30fps
Video decode 1080P@60fps
Ethernet Support RGMII / RMII Ethernet
Display MIPI DSI 4 Lane
RAM 2GB LPDDR4X 3733Mhz
Storge Optional 16GB EMMC (Boot from TF card by default)
IO pins DDR4 SODIMM 260P route all IO

Ext-board parameters

Ext-board
Camera Maixmum supported 3 cameras (Default connecting one camera):
1 MIPI 4-LANE + 2 MIPI 2-LANE
Display Maixmum supported 4-LANE MIPI DSI
SD card slot Default boot from SD card
Recording 2 onboard MEMS MIC support dual MIC noise reduction
Audio output Support 3.5MM interface, used for stereo output or MIC input
Network Support Gigabit Ethernet (ETH) and 2.4GHZ WI-FI
USB 1 x USB2.0HS used for OTG or HOST
1 x USB-UART
Others 1 User key，
3 x System keys，
1 x RTC，
2 x LEDs
Interface Reserved CJTAG，used for connecting corresponding debugger，
2 x 12 pin headers，
4 x M2.5 screw positioning holes

Hardware summary information

Camera comparison

  • Low-end camera gc4653 400w basic experience

  • High-end camera os04a10 400w enhanced night vision edition

How to change camera

Others

Visit Maix-III AXera-Pi (FAQ) if you have any trouble.