Generating Non-Stationary Textures using Self-Rectification. (arXiv:2401.02847v2 [cs.CV] UPDATED)

This paper addresses the challenge of example-based non-stationary texture
synthesis. We introduce a novel twostep approach wherein users first modify a
reference texture using standard image editing tools, yielding an initial rough
target for the synthesis. Subsequently, our proposed method, termed
"self-rectification", automatically refines this target into a coherent,
seamless texture, while faithfully preserving the distinct visual
characteristics of the reference exemplar. Our method leverages a pre-trained
diffusion network, and uses self-attention mechanisms, to gradually align the
synthesized texture with …

arxiv challenge cs.cv editing editing tools example image novel paper reference standard synthesis texture tools