Feature-Based Textures

Ganesh Ramanarayanan, Kavita Bala, Bruce Walter

Abstract: This paper introduces feature-based textures, a new image representation that combines features and samples for high-quality texture mapping. Features identify boundaries within an image where samples change discontinuously. They can be extracted from vector graphics representations, or explicitly added to raster images to improve sharpness. Texture lookups are then interpolated from samples while respecting these boundaries. We present results from a software implementation of this technique demonstrating quality, efficiency and low memory overhead.