Deep generative models in inversion: The impact of the generator's nonlinearity and development of a new approach based on a variational autoencoder

J Lopez-Alvis, E Laloy, F Nguyen, T Hermans - Computers & Geosciences, 2021‏ - Elsevier
When solving inverse problems in geophysical imaging, deep generative models (DGMs)
may be used to enforce the solution to display highly structured spatial patterns which are …

Deferred neural rendering: Image synthesis using neural textures

J Thies, M Zollhöfer, M Nießner - Acm Transactions on Graphics (TOG), 2019‏ - dl.acm.org
The modern computer graphics pipeline can synthesize images at remarkable visual quality;
however, it requires well-defined, high-quality 3D content as input. In this work, we explore …

On the continuity of rotation representations in neural networks

Y Zhou, C Barnes, J Lu, J Yang… - Proceedings of the IEEE …, 2019‏ - openaccess.thecvf.com
In neural networks, it is often desirable to work with various representations of the same
space. For example, 3D rotations can be represented with quaternions or Euler angles. In …

Deepvoxels: Learning persistent 3d feature embeddings

V Sitzmann, J Thies, F Heide… - Proceedings of the …, 2019‏ - openaccess.thecvf.com
In this work, we address the lack of 3D understanding of generative neural networks by
introducing a persistent 3D feature embedding for view synthesis. To this end, we propose …

Stochastic normalizing flows

H Wu, J Köhler, F Noé - Advances in neural information …, 2020‏ - proceedings.neurips.cc
The sampling of probability distributions specified up to a normalization constant is an
important problem in both machine learning and statistical mechanics. While classical …

Learning meaningful representations of protein sequences

NS Detlefsen, S Hauberg, W Boomsma - Nature communications, 2022‏ - nature.com
How we choose to represent our data has a fundamental impact on our ability to
subsequently extract information from them. Machine learning promises to automatically …

Normalizing flows on tori and spheres

DJ Rezende, G Papamakarios… - International …, 2020‏ - proceedings.mlr.press
Normalizing flows are a powerful tool for building expressive distributions in high
dimensions. So far, most of the literature has concentrated on learning flows on Euclidean …

CryoDRGN2: Ab initio neural reconstruction of 3D protein structures from real cryo-EM images

ED Zhong, A Lerer, JH Davis… - Proceedings of the …, 2021‏ - openaccess.thecvf.com
Protein structure determination from cryo-EM data requires reconstructing a 3D volume (or
distribution of volumes) from many noisy and randomly oriented 2D projection images. While …

Data augmentation in high dimensional low sample size setting using a geometry-based variational autoencoder

C Chadebec, E Thibeau-Sutre, N Burgos… - … on Pattern Analysis …, 2022‏ - ieeexplore.ieee.org
In this paper, we propose a new method to perform data augmentation in a reliable way in
the High Dimensional Low Sample Size (HDLSS) setting using a geometry-based …

Coordinate Independent Convolutional Networks--Isometry and Gauge Equivariant Convolutions on Riemannian Manifolds

M Weiler, P Forré, E Verlinde, M Welling - arxiv preprint arxiv:2106.06020, 2021‏ - arxiv.org
Motivated by the vast success of deep convolutional networks, there is a great interest in
generalizing convolutions to non-Euclidean manifolds. A major complication in comparison …