Diverse Texture

Diverse texture synthesis focuses on generating realistic and varied textures from limited input data, addressing challenges in areas like 3D character animation and material analysis. Current research emphasizes unsupervised learning techniques and novel architectures, including transformers and neural cellular automata, to improve both the quality and efficiency of texture generation. These advancements are impacting fields ranging from computer graphics and animation to material science, enabling more realistic simulations and automated texture design for various applications. The development of compact, efficient models is a key trend, facilitating broader accessibility and deployment.

Papers

March 25, 2024

Make-It-Vivid: Dressing Your Animatable Biped Cartoon Characters from Text
Junshu Tang, Yanhong Zeng, Ke Fan, Xuheng Wang, Bo Dai, Kai Chen, Lizhuang Ma
Text Modality Texture Synthesis 3D Character Diverse Texture

November 17, 2023

3D-TexSeg: Unsupervised Segmentation of 3D Texture using Mutual Transformer Learning
Iyyakutti Iyappan Ganapathi, Fayaz Ali, Sajid Javed, Syed Sadaf Ali, Naoufel Werghi
Mutual Learning Unsupervised Segmentation 3D G 3D Texture Diverse Texture

May 10, 2023

Learning in a Single Domain for Non-Stationary Multi-Texture Synthesis
Xudong Xie, Zhen Zhu, Zijie Wu, Zhiliang Xu, Yingying Zhu
LeArning Abstract Texture Synthesis Long Form Generation Single Domain Diverse Texture

February 23, 2022

U-Attention to Textures: Hierarchical Hourglass Vision Transformer for Universal Texture Synthesis
Shouchang Guo, Valentin Deschaintre, Douglas Noll, Arthur Roullier
Vision Transformer Hierarchical Attention Texture Synthesis Hierarchical Vision Transformer Efficient Attention Texture Quality Diverse Texture

November 26, 2021

$\mu$NCA: Texture Generation with Ultra-Compact Neural Cellular Automata
Alexander Mordvintsev, Eyvind Niklasson
Texture Synthesis Neural Cellular Automaton Texture Generation Procedural Model Diverse Texture

Diverse Texture

Papers

Make-It-Vivid: Dressing Your Animatable Biped Cartoon Characters from Text

3D-TexSeg: Unsupervised Segmentation of 3D Texture using Mutual Transformer Learning

Learning in a Single Domain for Non-Stationary Multi-Texture Synthesis

U-Attention to Textures: Hierarchical Hourglass Vision Transformer for Universal Texture Synthesis

$\mu$NCA: Texture Generation with Ultra-Compact Neural Cellular Automata