Texture Synthesis

Texture Synthesis on Near-regular and Irregular Textures

Introduction

We consider textures as different forms of departures from a regular texture pattern. Our previous work has focused on faithful texture synthesis for near-regular texture departing along the color and intensity axes while the underlying geometric regularity is well preserved. In this project, we address the issue of faithful texture synthesis for textures that have both the structural and color/ intensity deformations. The texture spectrum below shows several examples of regular, near-regular, irregular, and stochastic textures. We treat an irregular texture as a deformation from a regular one by deducing a deformation field between them. The novel view is to treat the deformation field itself as a texture that is both visual and functional. As a result, we can handle texture synthesis for a much larger variety of near-regular textures.

Publications

· Yanxi Liu and Wen-Chieh Lin, “Deformable Texture: the Irregular-Regular-Irregular Cycle,” The CMU technical report: CMU-RI-03-26, 2003.

· Yanxi Liu, Yanghai Tsin, and Wen-Chieh Lin, “The Promise and Perils of Near-Regular Texture,” International Journal of Computer Vision, 2003. In press.[paper][results]

· Yanxi Liu and Yanghai Tsin, “The Promise and Perils of Near-Regular Texture,” The Second International Workshop on Texture Analysis and Synthesis, 2002. [paper][results]

Near Regular Texture Synthesis Results

Motivated by the low structural fidelity for near-regular textures in current texture synthesis algorithms, we propose and implement an alternative texture synthesis method for near-regular texture. We view such textures as statistical departures from regular patterns and argue that a thorough understanding of their structures in terms of their translation symmetries can enhance existing methods of texture synthesis. We demonstrate the perils of texture synthesis for near-regular texture and the promise of faithfully preserving the regularity as well as the randomness in a near-regular texture sample.

The input texture.