3D Packaging Measurement via Computer Vision
A computer vision application that utilizes ArUco markers and a dual-camera setup to accurately measure the 3D dimensions of packaging boxes and visualize them in real-time.
This project is a specialized Computer Vision application designed to automate the physical measurement of 3D objects, specifically packaging boxes. By utilizing ArUco markers as a reliable physical scale reference, the system processes image feeds from two different camera angles to calculate the real-world length, width, and height of an object in centimeters. Built with a sleek Streamlit web interface, the application provides an easy-to-use dashboard for users to monitor the camera feeds and see the calculated dimensions instantly. Beyond just outputting numerical data (which is automatically saved to a JSON file), the system generates a 3D wireframe visualization of the detected box. Furthermore, it features an advanced integration with Blender via its Python API (bpy), allowing the system to automatically generate and render 3D models based on the measured physical dimensions.
Technologies Used
Key Features
- ArUco Marker Detection: Dynamically determines the pixel-to-centimeter ratio using standard ArUco markers.
- Dual-Camera 3D Measurement: Calculates precise Length, Width, and Height using multi-angle image processing.
- Interactive Streamlit UI: A clean, real-time web dashboard for operation and monitoring.
- Data Export: Automatically saves the detected measurements into structured JSON files.
- 3D Visualization & Blender Integration: Generates 3D wireframes and automates 3D box rendering in Blender using bpy.