Název: | Soft shadow computation using precomputed line space visibility information |
Autoři: | Keul, Kevin Klee, Nicolas Müller, Stefan |
Citace zdrojového dokumentu: | Journal of WSCG. 2017, vol. 25, no. 2, p. 97-106. |
Datum vydání: | 2017 |
Nakladatel: | Václav Skala - UNION Agency |
Typ dokumentu: | článek article |
URI: | wscg.zcu.cz/WSCG2017/!_2017_Journal_WSCG-No-2.pdf http://hdl.handle.net/11025/26287 |
ISSN: | 1213-6972 (print) 1213-6980 (CD-ROM) 1213-6964 (on-line) |
Klíčová slova: | vizualizace;počítačová grafika;sledování paprsku;datové struktury;algoritmy viditelnosti |
Klíčová slova v dalším jazyce: | visualization;computer graphics;ray tracing;data structures;visibility algorithms |
Abstrakt v dalším jazyce: | Shadows are one of the most important effects to create realism in rendering. Most real-time applications use some sort of image based technique like shadow mapping. While these techniques are quite fast, they often struggle at rendering realistic and accurate shadows of area lights. To produce correct shadows it is therefore often necessary to use ray tracing with some sort of acceleration method, nowadays mostly GPU based BVH which have their downsides in real-time rendering. We present a novel approach in calculating approximated but fast shadows using the line space as precomputed data structure for visibility information. With that it is possible to skip intersection tests with scene geometry and completely rely on the line space data structure for the shadow computations of area lights. Our approach is therefore almost scene-independent and is able to produce accurate shadows with better performance in comparison to typical ray tracing data structures. |
Práva: | © Václav Skala - UNION Agency |
Vyskytuje se v kolekcích: | Volume 25, Number 2 (2017) |
Soubory připojené k záznamu:
Soubor | Popis | Velikost | Formát | |
---|---|---|---|---|
Keul.pdf | Plný text | 2,75 MB | Adobe PDF | Zobrazit/otevřít |
Použijte tento identifikátor k citaci nebo jako odkaz na tento záznam:
http://hdl.handle.net/11025/26287
Všechny záznamy v DSpace jsou chráněny autorskými právy, všechna práva vyhrazena.