mindar 0.1.1

High-performance MindAR implementation for real-time image recognition on edge devices. Developed by FANSEE LAB.

MindAR Python

High-performance MindAR implementation for real-time image recognition on edge devices.

Features

• MindAR Compatible: Full compatibility with MindAR .mind file format
• High Performance: Optimized for Raspberry Pi and edge devices
• Real-time Detection: Efficient feature detection and matching
• Pure Python: No complex dependencies, easy deployment

Installation

cd mindar
pip install -e .

Usage

Basic Detection

import cv2
from mindar import Detector, Matcher, MindARCompiler

# Load target images and create detector
detector = Detector(width=640, height=480)
matcher = Matcher()

# Detect features in image
image = cv2.imread("target.jpg", cv2.IMREAD_GRAYSCALE)
result = detector.detect(image)
feature_points = result["feature_points"]

print(f"Detected {len(feature_points)} features")

Compile .mind Files

from mindar import MindARCompiler

compiler = MindARCompiler()

# Compile images to .mind file
success = compiler.compile_directory("./images", "./targets.mind")
if success:
    print("✅ Compilation successful")

# Load compiled targets
mind_data = compiler.load_mind_file("./targets.mind")

Performance

Optimized for edge devices like Raspberry Pi 4:

• Detection: ~50ms per frame (640x480)
• Matching: ~20ms per target
• Memory: <100MB usage

License

MIT License - Compatible with original MindAR project