Skip to content

README.md

Latest commit

 

History

History
48 lines (33 loc) · 1.6 KB

README.md

File metadata and controls

48 lines (33 loc) · 1.6 KB

3D Lane Labeling Machine

The 3D Lane Labeling Machine is a tool designed for autonomous driving applications, enabling efficient labeling of 3D lane data. This tool helps create training datasets for autonomous vehicle perception systems by providing an intuitive interface for labeling lane markings in 3D space.

🚀 Overview

Features include:

  • 3D visualization of camera images and LiDAR point clouds
  • Interactive lane point labeling
  • Support for OpenLane V1 data format
  • Camera and LiDAR calibration support
  • Easy-to-use graphical interface

🔨 Installation

pip install -r requirements.txt

▶️ Execution

python main.py

📄 Documentation

This version supports the OpenLane V1 format and requires images, PCD files, and camera and lidar calibration files.

Data

  • Put your images in data/image/image folder.
  • Put your pcd files in data/image/pcd folder.
  • Put your calibration files in calibration folder. (r.txt, t.txt, k.txt, distortion.txt)

Usage

  1. Run python main.py
  2. Choose the class of lane.
  3. Click the point you want to label on the left image or the right window.
  4. Click the 'Add Lane' button. Also, you can delete the points and lane by clicking the 'Delete' button.
  5. Click the 'Save' button to save the label or 'Next' button to save and move to the next image. (data/label folder)
  • You can use keyboard shortcuts to add, delete, move and so on..

This project is based on: Road_Labeler