Old School Procedural Texture Extraveganza

 This is an awesome talk by procedural artist Anastasia Opara of Embark Studios titled "More Like This, Please! Single Example Texture Synthesis and Remixing".  She starts out by trying to give an intuitive feel for what is going on in texture synthesis algorithms.  We then go on a wild ride through the history of texture synthesis.  She then builds on top of that research for her unique approach.

The talk also discusses how the notion of texture synthesis can be integrated into a digital artist's workflow.  Some interesting 'old school' approaches to style transfer are also discussed based on texture synthesis algorithms.

The 'funky style-transfer' approach using the Levoy paper algorithm is pretty slick.

So is the 3d rock synthesis example (which she really quickly mentions in passing) based on incorporating displacement maps into the texture synthesis procedure.

Anastasia's MultiResolution Random Order algorithm is based on the Paul Harrison pdf thesis 'Image Texture Tools' available here.

Michael Ashikkhmin's paper on 'Synthesizing Nature Textures' is available here.

The Wei et al. paper on 'Fast Texture Synthesis Using Tree-Structured Vector Quantization' is available here.

The original 'Texture Synthesis by Non-parametric Sampling' paper by Efos and Leung is available here.

It's great that Anastasia gave tribute to all of the texture synthesis work that came before her (not always true with some neural style transfer work).

Here's a paper on 'Appearance Space Texture Synthesis' that was not discussed in the talk, but has some relevance.

So of course i would love to see all of this available in PyTorch implementations.


Popular posts from this blog

Simulating the Universe with Machine Learning

CycleGAN: a GAN architecture for learning unpaired image to image transformations

Pix2Pix: a GAN architecture for image to image transformation