When you start a new pipeline that involves geometry processing algorithms, choosing the right one can be challenging. With numerous options, some well-known and others less so, it is common to invest time in implementing an algorithm only to find out that it either does not work or falls short of expectations. The issue may not lie with the algorithm itself! In fact, it could be a great algorithm but not well-tuned for your specific use cases. Alternatively, it might be well-suited but challenging and time-consuming to implement. So, how do you find the right version that is both effective for your problem and quick and simple to implement?

​

Some algorithms can be found in libraries. Some other have never been code-published. But even having it in a library may not be efficient for a business because the solution is expensive and hard to integrate.

​

Drawing on my experience, I have encountered various use cases and identified suitable geometry processing algorithms both as part of libraries and only paper-published to address those cases effectively. While I may not have exhaustive knowledge, I often have a good starting point for quick and efficient solutions to common problems in geometry processing.