STS-GAN: Can We Synthesize Solid Texture with High Fidelity from Arbitrary 2D Exemplar?

TL;DR

This paper introduces STS-GAN, a novel generative adversarial network framework that synthesizes high-fidelity 3D solid textures from 2D exemplars, addressing limitations of previous methods in accurately learning arbitrary textures.

Contribution

The paper presents a new GAN-based framework that effectively extends 2D textures to 3D solid textures with high fidelity, using multi-scale discriminators for improved realism.

Findings

Generates high-fidelity solid textures matching 2D exemplars

Uses multi-scale discriminators to evaluate texture similarity

Outperforms existing methods in texture synthesis quality

Abstract

Solid texture synthesis (STS), an effective way to extend a 2D exemplar to a 3D solid volume, exhibits advantages in computational photography. However, existing methods generally fail to accurately learn arbitrary textures, which may result in the failure to synthesize solid textures with high fidelity. In this paper, we propose a novel generative adversarial nets-based framework (STS-GAN) to extend the given 2D exemplar to arbitrary 3D solid textures. In STS-GAN, multi-scale 2D texture discriminators evaluate the similarity between the given 2D exemplar and slices from the generated 3D texture, promoting the 3D texture generator synthesizing realistic solid textures. Finally, experiments demonstrate that the proposed method can generate high-fidelity solid textures with similar visual characteristics to the 2D exemplar.