3d classification


I’m interested in analyzing/classifying 3d shapes.

Please find two examples attached below.

The 3d geometries you see above are not ‘hand-made’ but generated from generative systems/algorithmic workflows, coded through 3d api offered by 3d modeling/simulation programmes.

Q1. Could anyone please suggest any specific libraries/ online courses to learn so one can identify and cluster sub-geometries automatically? Are there any deep learning methodologies/ ML approaches in general, which can learn from a limited data-set and not require to be pre-trained?

I have heard about Keras, and also am seeing that TensorFlow and PyTorch are now slowly expanding to the 3d space. To better place myself, I have completed courses in Linear Algebra, Multivariate Calculus, and Data Science this year, but am not clear on the path to follow ahead.

Q2. I come from a geometry automation background in the field of design and manufacturing. I’m interested to see if there are any opportunities for people with strong 3d scripting/ visual geometry background to contribute positively in the field of data science.

Will be great to hear your thoughts on this. Any advice/ pointers will be greatly appreciated!

Thank you!

1 Like