Jun 2, 2020 · This paper is concerned with a fundamental problem in geometric deep learning that arises in the construction of convolutional neural networks on surfaces.

Aug 12, 2020 · We propose a network architecture for surfaces that consists of vector-valued, rotation-equivariant features.

Code for Harmonic Surface Networks, an approach for deep learning on surfaces operating on vector-valued, rotation-equivariant features.

We propose a network architecture for surfaces that consists of vector-valued, rotation-equivariant features.

This paper is concerned with a fundamental problem in geometric deep learning that arises in the construction of convolutional neural networks on surfaces.

Jun 2, 2020 · We introduce Harmonic Surface Networks, which combine a vector-valued convolution operation on surfaces and rotation- equivariant filter kernels ...

Features expressed as complex vectors. ○ Use circular harmonics (harmonic networks: learn radial and angular parameters). ○ rotational-equivariant kernels.

Duration: 19:47
Posted: Aug 24, 2020

A network architecture for surfaces that consists of vector-valued, rotation-equivariant features that makes it possible to locally align features, ...

Duration: 10:58
Posted: Apr 24, 2021