Convert videos into 3D point clouds using COLMAP.
This project extracts frames from a video, runs COLMAP to create a point cloud, and saves it in PLY format. The resulting point cloud can be imported into anything you need for visualization and further manipulation.
"I've been keeping a journal to provide more context as I learn how to make this." - zbuhrer
I'll elaborate on these steps in the future, but the quick 'n easy way to get this set up is as follows:
- Install COLMAP, referring to their instructions.
- Clone this project from Git:
git clone https://github.com/zbuhrer/pointcloud_converter
- Create and activate a virtual environment (Linux/BSD environments):
python3 -m venv venv source venv/bin/activate - Install dependencies (while inside
venv!):pip3 install -r requirements.txt
- Put your mp4 video file in the root of this project (rename it to
video_input.mp4) — refer to this documentation for best practices. - Start the application:
python3 .
Your terminal window will produce a lot of output for several hours. Refer to the roadmap below if you think of something this isn't doing yet.
- Let the user select filters to apply or ignore
- Better logging/console output so it's more clear what it's doing
- Hey how about a loading bar at least?
- System requirements documentation
- Need to work on a table of stats for models that I've extracted so they can be studied
- UI?
- Should this be a console-only app? I've never made a GUI before, that could be fun!
This project is licensed under the Apache License 2.0 - see the LICENSE file for details.