Jigsaw image mosaics

Abstract

This paper introduces a new kind of mosaic, called Jigsaw Image Mosaic (JIM), where image tiles of arbitrary shape are used to compose the final picture. The generation of a Jigsaw Image Mosaic is a solution to the following problem: given an arbitrarily-shaped container image and a set of arbitrarily-shaped image tiles, fill the container as compactly as possible with tiles of similar color to the container taken from the input set while optionally deforming them slightly to achieve a more visually-pleasing effect. We approach the problem by defining a mosaic as the tile configuration that minimizes a mosaicing energy function. We introduce a general energy-based framework for mosaicing problems that extends some of the existing algorithms such as Photomosaics and Simulated Decorative Mosaics. We also present a fast algorithm to solve the mosaicing problem at an acceptable computational cost. We demonstrate the use of our method by applying it to a wide range of container images and tiles.

Perceptually inspired real-time artistic style transfer for video stream

“…These approaches include color transfer [4,21,23,24] which is a color correction method for borrowing one images color characteristics from another, photomosaic [13][14][15]22] which represents a target image by assembling given photographic examples, collage [12]. In this study, we concentrate mainly on the style transfer approach.…”

Exploring Swarm-Based Visual Effects

“…In its simplest form, an image can be easily converted into photomosaic of square tiles. In a more sophisticated manner, various components in an image can be segmented out first before generating mosaic patterns for each of them [6,7]. The latter process produces a much more stylistic tile composition.…”

Interactive by-example design of artistic packing layouts

“…Seeking to place samples that evaluate a function such that aliasing is minimized, Mitchell [1987] argued that samples should have "blue noise" characteristics, that is: the distance to the neighbors should not be smaller than a threshold. Placing primitives for artistic purposes in 2D is widely used for non-photorealistic rendering, e. g., for stippling [Deussen et al 2000;Hiller et al 2003], mosaics [Hausner 2001;Kim and Pellacini 2002] or texture synthesis [Lagae and Dutré 2005]. In particular Hiller et al [2003] who distributes primitives in the plane such that they follow a prescribed density, is a similar case of our system that produces distributions that follow rules inferred from the users' interaction with the distribution itself.…”