國立高雄大學圖資館 |

登入

回首頁

GenShade :An evolutionary approach to automatic and interactive procedural texture generation.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	GenShade :
其他題名:	An evolutionary approach to automatic and interactive procedural texture generation.
作者:	Ibrahim, Alaa Eldin M.
面頁冊數:	120 p.
附註:	Chair: Donald H. House.
附註:	Source: Dissertation Abstracts International, Volume: 59-12, Section: B, page: 6379.
Contained By:	Dissertation Abstracts International59-12B.
標題:	Computer Science.
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=9915243
ISBN:	0599139633

GenShade :An evolutionary approach to automatic and interactive procedural texture generation.
Ibrahim, Alaa Eldin M.

GenShade :An evolutionary approach to automatic and interactive procedural texture generation. [electronic resource] - 120 p.

Chair: Donald H. House.

Thesis (Ph.D.)--Texas A&M University, 1998.

The current system operates in both artist directed and automatic texture matching modes. In interactive artist directed mode, aesthetic evaluation and selection is controlled by the user. In automatic mode, a multiple-generation population is used that is updated by reproduction every generation. This procedure includes the notions of aging and gender, and runs in a multiple processor mode. The system evaluates generated textures by comparing them with target images of a desired texture, using a multiresolution image querying metric. Examples are shown of results of this method in generating textures over a wide range of texture complexities using both the automatic and interactive modes.

ISBN: 0599139633Subjects--Topical Terms:

212513
Computer Science.

GenShade :An evolutionary approach to automatic and interactive procedural texture generation.
LDR:02918nmm 2200277 450 001 155163
005 20030117152117.5
008 230606s1998 eng d
020 $a 0599139633
035 $a 00087688
035 $a 155163
040 $a UnM $c UnM
100 0 $a Ibrahim, Alaa Eldin M. $3 212864
245 1 0 $a GenShade : $b An evolutionary approach to automatic and interactive procedural texture generation. $h [electronic resource]
300 $a 120 p.
500 $a Chair: Donald H. House.
500 $a Source: Dissertation Abstracts International, Volume: 59-12, Section: B, page: 6379.
502 $a Thesis (Ph.D.)--Texas A&M University, 1998.
520 # $a The current system operates in both artist directed and automatic texture matching modes. In interactive artist directed mode, aesthetic evaluation and selection is controlled by the user. In automatic mode, a multiple-generation population is used that is updated by reproduction every generation. This procedure includes the notions of aging and gender, and runs in a multiple processor mode. The system evaluates generated textures by comparing them with target images of a desired texture, using a multiresolution image querying metric. Examples are shown of results of this method in generating textures over a wide range of texture complexities using both the automatic and interactive modes.
520 # $a This study describes the development of a new methodology for generating procedural textures that uses enhanced genetic algorithm techniques. This method solves problems associated with previous methods for texture generation in that it is able to generate non-homogenous and structurally complex textures as well as homogenous and simple ones. It is especially successful in producing large families of textures that are similar to target textures, allowing artists a wide variety of choices and control tools.
520 # $a To experiment with this methodology, the GenShade software system was developed as a testing platform. The system used existing research results in the areas of texture synthesis and genetic algorithms as a starting point. Since then, it has evolved, over more than two years, in order to improve its performance by inventing new concepts. A process involving informal experimentation was used to test the new ideas and to decide whether to pursue them further or to select other new directions. This process continued until the system was consistently generating procedural textures similar to target textures. Finally, a series of formal experiments was conducted in order to verify this methodology.
590 $a School code: 0803.
650 # 0 $a Computer Science. $3 212513
650 # 0 $a Fine Arts. $3 212421
650 # 0 $a Architecture. $3 208437
710 0 # $a Texas A&M University. $3 212471
773 0 # $g 59-12B. $t Dissertation Abstracts International
790 $a 0803
790 1 0 $a House, Donald H., $e advisor
791 $a Ph.D.
792 $a 1998
856 4 0 $u http://libsw.nuk.edu.tw/login?url=http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=9915243 $z http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=9915243