In graph neural network frameworks, there is always a template with a shared structure among all graphs. I have meshes that are registered but obviously, Lalpalcian and their geometry are different. How can we define Laplacian filters for each meshes and not being based on a fixed graph/mesh. In ChebNet, for instance, the structure is shared and coarsening is done beforehand since the graphs are all the same but having different node features. In meshes, this is not the case. That being said, is there any method to use Laplace Beltrami Operator for each mesh and then use them for convolution layers?