{"title":"交互体素化极面阴影体","authors":"Chris Wyman","doi":"10.1145/1899950.1900003","DOIUrl":null,"url":null,"abstract":"Current algorithms for rendering shadows inside participating media hit a bottleneck when computing light visibility throughout the media. These algorithms rely either on sampling along viewing rays, often thrashing memory caches, or slower analytic solutions using object-space computations, such as shadow volumes. We present a new cache-coherent sampling technique that computes volumetric light visibility that requires as little as one texture lookup per pixel. We efficiently voxelize shadow volumes in epipolar-space using standard parallel scan operations. The only step dependent on geometric complexity is an image-space voxelization [Eisemann and Décoret 2006] that often takes under a millisecond. This allows us to render shadows in participating media at up to 300 frames per second.","PeriodicalId":354911,"journal":{"name":"ACM SIGGRAPH ASIA 2010 Sketches","volume":"18 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2010-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"5","resultStr":"{\"title\":\"Interactive voxelized epipolar shadow volumes\",\"authors\":\"Chris Wyman\",\"doi\":\"10.1145/1899950.1900003\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Current algorithms for rendering shadows inside participating media hit a bottleneck when computing light visibility throughout the media. These algorithms rely either on sampling along viewing rays, often thrashing memory caches, or slower analytic solutions using object-space computations, such as shadow volumes. We present a new cache-coherent sampling technique that computes volumetric light visibility that requires as little as one texture lookup per pixel. We efficiently voxelize shadow volumes in epipolar-space using standard parallel scan operations. The only step dependent on geometric complexity is an image-space voxelization [Eisemann and Décoret 2006] that often takes under a millisecond. This allows us to render shadows in participating media at up to 300 frames per second.\",\"PeriodicalId\":354911,\"journal\":{\"name\":\"ACM SIGGRAPH ASIA 2010 Sketches\",\"volume\":\"18 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2010-12-15\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"5\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"ACM SIGGRAPH ASIA 2010 Sketches\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/1899950.1900003\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"ACM SIGGRAPH ASIA 2010 Sketches","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/1899950.1900003","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 5
摘要
当前用于渲染参与媒体内部阴影的算法在计算整个媒体的光可见性时遇到瓶颈。这些算法要么依赖于沿着观察光线的采样,通常是抖动内存缓存,要么依赖于使用对象空间计算(如阴影体积)的较慢的分析解决方案。我们提出了一种新的缓存相干采样技术,该技术可以计算每像素只需要一次纹理查找的体积光可见性。我们使用标准的并行扫描操作有效地对极空间中的阴影体素化。唯一依赖于几何复杂性的步骤是图像空间体素化[Eisemann and dsamcoret 2006],通常需要不到一毫秒的时间。这允许我们以每秒300帧的速度渲染参与媒体中的阴影。
Current algorithms for rendering shadows inside participating media hit a bottleneck when computing light visibility throughout the media. These algorithms rely either on sampling along viewing rays, often thrashing memory caches, or slower analytic solutions using object-space computations, such as shadow volumes. We present a new cache-coherent sampling technique that computes volumetric light visibility that requires as little as one texture lookup per pixel. We efficiently voxelize shadow volumes in epipolar-space using standard parallel scan operations. The only step dependent on geometric complexity is an image-space voxelization [Eisemann and Décoret 2006] that often takes under a millisecond. This allows us to render shadows in participating media at up to 300 frames per second.