💡 Get 7x PDF for 3D Data Tutorials here: https://learngeodata.eu/3d-newsletter/
This tutorial gives a detailed workflow for labeling point clouds, particularly focusing on geospatial applications with large datasets. I outline a semi-automated approach that leverages tools like CloudCompare and unsupervised learning algorithms to create labeled datasets for training 3D deep learning models.
Here are some of the topics I cover in this video:
✅ How to organize and process point clouds, including using voxel grids, connected components, and thresholding to efficiently analyze and manage data.
✅ How to use semi-automated techniques to classify points into categories such as ground, walls, and wires, with a focus on precision to avoid creating a "garbage dataset."
✅ Possibilities for fully automating the process using Python
This video is a great resource for anyone who wants to learn how to build 3D data tools.
🙋 FOLLOW ME
Linkedin: https://www.linkedin.com/in/florent-poux-point-cloud/
Medium: https://medium.com/@florentpoux
WHO AM I?
If we haven’t yet before - Hey 👋 I’m Florent, a professor-turned-entrepreneur, and I’ve somehow become one of the most-followed 3D experts. Through my videos here on this channel and my writing, I share evidence-based strategies and tools to help you be better coders and 3D innovators.
📗 CHAPTERS
[00:00]: Introduction to Point Cloud Labeling Workflow
[01:16]: Improving Point Cloud Visibility and Color
[02:35]: Defining and Managing Point Cloud Classes
[04:00]: Analyzing Point Cloud with Various Tools
[06:50]: CSF Filter for Ground and Off-Ground Points
[08:07]: Organizing Point Cloud Files and Backup
[09:40]: Ransac Shape Detection for Planes
[13:22]: Connected Component and Remaining Elements
[18:05]: Labeling Walls and Wires with Cloud Layer
[25:21]: Finalizing and Exporting Classified Point Cloud