MaixPy Development Skill

MaixPy v4 is a Python SDK for edge AI development on Sipeed hardware. This skill provides patterns, examples, and best practices.

Hardware Comparison

Quick Start

from maix import camera, display, image, nn, app

detector = nn.YOLOv8(model="/root/models/yolov8n.mud", dual_buff=True)
cam = camera.Camera(detector.input_width(), detector.input_height(), detector.input_format())
disp = display.Display()

while not app.need_exit():
    img = cam.read()
    objs = detector.detect(img, conf_th=0.5, iou_th=0.45)
    for obj in objs:
        img.draw_rect(obj.x, obj.y, obj.w, obj.h, color=image.COLOR_RED)
        img.draw_string(obj.x, obj.y, f'{detector.labels[obj.class_id]}: {obj.score:.2f}')
    disp.show(img)

Reference Documentation

Device Detection

from maix import sys

device_id = sys.device_id()  # "maixcam", "maixcam2"

if device_id == "maixcam2":
    model = "/root/models/yolo11s.mud"  # Larger model
else:
    model = "/root/models/yolov8n.mud"  # Nano model

Model Paths

Pre-installed models in /root/models/:

• Detection: yolov8n.mud, yolo11n.mud, yolo11s.mud
• Segmentation: yolo11nseg.mud, yolov8nseg.mud
• Pose: yolo11npose.mud, yolov8npose.mud
• Face: yolov8n_face.mud, retinaface.mud
• Classifier: mobilenetv2.mud
• OCR: pp_ocr.mud
• Hand: hand_landmarks.mud

App Development

my_app/
├── app.yaml      # Config (see assets/app.yaml.template)
├── main.py       # Entry point
└── icon.png      # App icon (128x128)

Resources

• Docs: https://wiki.sipeed.com/maixpy/
• API: https://wiki.sipeed.com/maixpy/api/index.html
• GitHub: https://github.com/sipeed/MaixPy
• Examples: https://github.com/sipeed/MaixPy/tree/main/examples
• Projects: https://github.com/sipeed/MaixPy/tree/main/projects
• MaixHub: https://maixhub.com
• Community: QQ群 862340358, t.me/maixpy