Point-cloud-3D-tree-reconstruction

💥Please Give a proper citation/clarification to this GitHub page and mentioning my Name if you're using my code/software for your paper. I have always noticed that some people directly use my code and then claim they “developed” or “proposed” the technique themselves. This is unacceptable! If you use this work in your research, please clearly acknowledge the source and cite this repository appropriately. Every time I encounter a paper that uses my work without proper attribution or citation, I will formally report it to the journal editors.

Quick Start.

This is a program for reconstructing a 3D tree model from scanned point clouds. The program is fully developed in C++

🎞️ Demo video:

We recommend all users first watching this [Demo Video] to understand its usage quickly. (Strongly Suggest!🔥🔥)

https://github.com/user-attachments/assets/1e220954-949b-4f4e-b950-efc51ec923c9

🪴 Software:

We have released the exe program and you can Download here. The users can directly run the program on Windows PCs without any configuration or compilation. We have sucessfully tested it for Win10 and Win11.

Here are also the example Point-cloud Files that I used in the demo video. You can download them for a quick start.

(1) Usage Instructions:

• Please first download the entire [folder], then double-click the TreeFromPoints.exe to execute the program.
• Then, press the Load Point Data button to load a point cloud file from your local disk. Example point data is available here.
• Finally, sequentially press the seven buttons from (1) Remove noises to (7) Optimize, and you will see the final 3D tree models in the right display panel.

(2) How to export the 3D models?

Actually, the 3D results are automatically saved under the exe folder path with filename as bark_texture.obj. Please refer to this issue for more details.

The following figures shows an example of an exported result which is opened by 3D Viewer.

♻️ Code:

For general use, we recommend using the exe program introduced above, which is very simple to use and there is no need for any configuration steps.

But if you want to customize the program for your own purposes, please Download the Source Code here. Our program is implemented in C++ codes, so you can compile with any C++ IDEs.

More Experimental Results.

License

This project is licensed under the MIT License. If you use this project, please clearly credit the original source by including the project name and repository link.

Copyright (c) 2026 Zhihao Liu