Multiscale imagery often merges several sources with differing appearance. For instance, Internet-based maps combine satellite and aerial photography. Zooming within these maps may reveal jarring transitions. We present a scheme that creates a visually smooth mipmap pyramid from stitched imagery at several scales. The scheme involves two new techniques. The first, structure transfer, is a nonlinear operator that combines the detail of one image with the local appearance of another. We use this operator to inject detail from the fine image into the coarse one while retaining color consistency. The improved structural similarity greatly reduces inter-level ghosting artifacts. The second, clipped Laplacian blending, is an efficient construction to minimize blur when creating intermediate levels. It considers the sum of all inter-level image differences within the pyramid. We demonstrate continuous zooming of map imagery from space to ground level.

Example-based texture synthesis algorithms generate novel texture images from example data. A popular hierarchical pixel-based approach uses spatial jitter to introduce diversity, at the risk of breaking coarse structure beyond repair. We propose a multiscale descriptor that enables appearance-space jitter, which retains structure. This idea enables repurposing of existing texture synthesis implementations for a qualitatively different problem statement and class of inputs: generating hybrids of structured images.

Example-based texture synthesis algorithms have gained widespread popularity for their ability to take a single input image and create a perceptually similar non-periodic texture. However, previous methods rely on single input exemplars that can capture only a limited band of spatial scales. For example, synthesizing a continent-like appearance at a variety of zoom levels would require an impractically high input resolution. In this paper, we develop a multiscale texture synthesis algorithm. We propose a novel example-based representation, which we call an exemplar graph, that simply requires a few low-resolution input exemplars at different scales. Moreover, by allowing loops in the graph, we can create infinite zooms and ifinitely detailed textures that are impossible with current example-based methods. We also introduce a technique that ameliorates inconsistencies in the user's input, and show that the application of this method yields improved interscale coherence and higher visual quality. We demonstrate optimizations for both CPU and GPU implementations of our method, and use them to produce animations with zooming and panning at multiple scales, as well as static gigapixel-sized images with features spanning many spatial scales.

Filtering is critical for representing image-based detail, such as textures or normal maps, across a variety of scales. While mipmapping textures is commonplace, accurate normal map filtering remains a challenging problem because of nonlinearities in shading--we cannot simply average nearby surface normals. In this paper, we show analytically that normal map filtering can be formalized as a spherical convolution of the normal distribution function (NDF) and the BRDF, for a large class of common BRDFs such as Lambertian, microfacet and factored measurements. This theoretical result explains many previous filtering techniques as special cases, and leads to a generalization to a broader class of measured and analytic BRDFs. Our practical algorithms leverage a significant body of previous work that has studied lighting-BRDF convolution. We show how spherical harmonics can be used to filter the NDF for Lambertian and low-frequency specular BRDFs, while spherical von Mises-Fisher distributions can be used for high-frequency materials.