nice works, but the theory is too hard, there is any mthods to understand the paper or methods?

[Asherchi]

No description provided.

[georg-bn]

What is your background?

[Asherchi]

I am a practitioner in 3D reconstruction. Understand some knowledge related to geometry and artificial intelligence.

[georg-bn]

I would recommend this course by Erik Bekkers to get an overview of equivariant methods in deep learning https://uvagedl.github.io/ . What we do in this paper is simpler than many of the methods presented there, but the underlying goal of equivariance is the same.

If you want a book reference for the math itself, with calculus and linear algebra as prerequisites, I would recommend "Groups and Symmetries" by Kosmann-Schwarzbach, link.

[Asherchi]

Thank you very much for your suggestion. I will study the relevant content carefully.