Learning a Neural 3D Texture Space From 2D Exemplars

By Philipp Henzler, Niloy J. Mitra, Tobias Ritschel. 2019

95 citations recorded by the source corpus.

We suggest a generative model of 2D and 3D natural textures with diversity, visual fidelity and at high computational efficiency. This is enabled by a family of methods that extend ideas from classic stochastic procedural texturing (Perlin noise) to learned, deep, non-linearities. Our model encodes all exemplars from a diverse set of textures without a need to be re-trained for each exemplar. Applications include texture interpolation, and learning 3D textures from 2D exemplars.

Subfields: Computational Mechanics, Computer Graphics and Computer-Aided Design, Computer Vision and Pattern Recognition, Generative Models

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

1. Jonathan T. Barron
2. Ben Mildenhall
3. Pratul P. Srinivasan
4. Matthew Tancik
5. Ricardo Martin-Brualla
6. Noah Snavely
7. Thomas A. Funkhouser
8. Andreas Geiger

In Computer Graphics and Computer-Aided Design

1. Jonathan T. Barron
2. Ben Mildenhall
3. Pratul P. Srinivasan
4. Matthew Tancik
5. Andreas Geiger
6. Thomas A. Funkhouser
7. Ricardo Martin-Brualla
8. Peter Hedman

In Computer Vision and Pattern Recognition

1. Jonathan T. Barron
2. Andreas Geiger
3. Pratul P. Srinivasan
4. Ben Mildenhall
5. Luc Van Gool
6. Jitendra Malik
7. Andrea Vedaldi
8. Ming-Hsuan Yang

In Generative Models

1. Jonathan T. Barron
2. Ben Mildenhall
3. Pratul P. Srinivasan
4. Matthew Tancik
5. Ricardo Martin-Brualla
6. Ravi Ramamoorthi
7. Thomas A. Funkhouser
8. Noah Snavely

Citations in the corpus

• Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domains (2020, background): “…volume density [27], occupancy [24], and signed distance [32], and have achieved state-of-the-art results across a variety of tasks such as shape representation [9, 10, 12, 13, 17, 26, 32], texture synthesis [15, 31], shape inference from images [22, 23], and novel view synthesis [27, 29, 35, 37].”
• IBRNet: Learning Multi-View Image-Based Rendering (2021, methodology): “Recent methods leverage convolutional neural networks (CNNs) to predict volumetric representations stored in voxel grids [20, 26, 27, 42, 59] or multi-plane images (MPIs) [12, 13, 31, 38, 54, 65].”