This project is written specifically for parsing .STEP files and converting them into a much easier to work with .JSON format.
As part of my research, I needed a tool that could read .STEP files and convert them into something easier to work with like a .JSON file. This was super helpful for the dataset generation and analysing the statistics of the 1 million .STEP files present in the ABC Dataset.
python, pythonOCC
Built with
Reading .STEP files and converting them into easier to manage formats like .JSON is extremely important for research in Geometric Deep Learning. When I started this, I could not find a lot of resources available for working with pythonOCC and the Open Cascade documentation was too vast to cover for a single-use project like this. I hope this provides some kind of template for anyone who has to work with large 3D datasets.
- The tool allows for easy conversion of a .STEP file into a .JSON file format.
- In addition to basic surface information, there is also a function that takes care of holes present in the object's surface.
For installation of the pythonOCC framework details, please refer https://github.com/tpaviot/pythonocc-core.