https://github.com/pauldanielml/point-cloud-shape-approximation

Using k-order Markov Path Optimization to fit shapes to sparse point cloud data.
https://github.com/pauldanielml/point-cloud-shape-approximation

computer-vision optimization point-cloud python robotics

Last synced: 22 days ago
JSON representation

Using k-order Markov Path Optimization to fit shapes to sparse point cloud data.

• Host: GitHub
• URL: https://github.com/pauldanielml/point-cloud-shape-approximation
• Owner: PaulDanielML
• Created: 2020-06-22T18:47:28.000Z (over 4 years ago)
• Default Branch: master
• Last Pushed: 2020-07-21T15:17:37.000Z (over 4 years ago)
• Last Synced: 2024-11-09T18:48:56.154Z (3 months ago)
• Topics: computer-vision, optimization, point-cloud, python, robotics
• Language: Python
• Homepage:
• Size: 492 KB
• Stars: 3
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Approximating shapes, positions and sizes from point cloud data for robotic grasping

Author: Paul Daniel ([email protected])

Approximation process	           |  Console output

:---------------------------------:|:-------------------------:

![](/media/gif_1.gif "Shape approximation")  |  ![](/media/img_1.png "Output")

YouTube Video:

[![video](https://img.youtube.com/vi/LC4D6ko-C4c/0.jpg)](https://www.youtube.com/watch?v=LC4D6ko-C4c)  

The following pseudocode shows the structure of the created algorithm used for finding the approximations. 

![](/media/algo.png "Algorithm")

Final project for the TU Berlin course "AI & Robotics: Lab Course". 

To reproduce set up rai as described [here](https://github.com/MarcToussaint/robotics-course#setup-for-robotics-practical-in-simulation). 

After successful setup, switch to the folder *course3-Simulation*:

    cd robotics-course/course3-Simulation

clone the this repo there: 

    git clone https://github.com/PaulDanielML/Point-Cloud-Object-Classification.git