Pattern Synthesis for Low-Resolution Displays

低分辨率显示器的图案合成

基本信息

批准号：
536352-2018
负责人：
Mould, David
金额：
$ 1.82万
依托单位：
Carleton University
依托单位国家：
加拿大
项目类别：
Engage Grants Program
财政年份：
2018
资助国家：
加拿大
起止时间：
2018-01-01 至 2019-12-31
项目状态：
已结题

来源：
https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=660350
关键词：
Pattern Synthesis Low Resolution Displays

项目摘要

Limbic Media creates interactive art installations in which arrays of lights react to sound in the environment, changing their color and brightness in response. At present, a lighting designer constructs patterns manually at considerable effort and expense. This project will investigate algorithms for automated pattern synthesis, making interesting visual structures automatically; these can either be used directly in an installation or serve as inspiration for the designer.****The light array has moderately low resolution, involving hundreds to tens of thousands of lights; contrast this with general photographic and synthetic imagery, where it is common to use millions of pixels. Low-resolution pattern synthesis has been little studied, although ventures into stippling, pixel art, and low-resolution colorings provide some precedent. Thus we have the research challenge of finding algorithms and data structures for creating and representing patterns in the low-resolution context.****Our first step will be to impose some structure on the arrangement of lights. Each light will be linked with a small number of its neighbours. Using these links, we can reason about the local structure and construct patterns. We will investigate abstract patterns, dynamic patterns, and even potentially representational patterns, that convey the sense of a recognizable scene despite the small number of lights available. These patterns will then be used by our industrial partner in creating breathtaking interactive lightshows and art.******

Limbic Media 创造了互动艺术装置，其中的灯光阵列对环境中的声音做出反应，从而改变其颜色和亮度。目前，照明设计师需要花费相当大的精力和费用来手动构建图案。该项目将研究自动模式合成的算法，自动制作有趣的视觉结构；这些可以直接在装置中使用，也可以作为设计师的灵感。****灯光阵列的分辨率较低，涉及数百到数万个灯光；将此与一般摄影和合成图像形成对比，后者通常使用数百万像素。尽管对点画、像素艺术和低分辨率着色的尝试提供了一些先例，但低分辨率图案合成的研究还很少。因此，我们面临的研究挑战是找到在低分辨率环境中创建和表示模式的算法和数据结构。****我们的第一步将是在灯光的排列上强加一些结构。每个灯都将与其少数邻居相连。使用这些链接，我们可以推断局部结构并构建模式。我们将研究抽象图案、动态图案，甚至潜在的代表性图案，尽管可用的灯光数量很少，但它们仍能传达可识别场景的感觉。然后，我们的工业合作伙伴将使用这些图案来创造令人惊叹的互动灯光秀和艺术。*****