DocumentCode :
381964
Title :
Feature-guided painterly image rendering
Author :
Li, Nan ; Huang, Zhiyong
Author_Institution :
Dept. of Comput. Sci., Nat. Univ. of Singapore, Singapore
Volume :
1
fYear :
2002
fDate :
2002
Abstract :
Non-photo-realistic rendering (NPR) refers to any technique which can produce a non-photo-realistic image. We present a method for automatically generating a stroke-based painting from a digital image. The rendering process generates rectangular brush strokes with suitable location, orientation and size. Inspired by the real painting process, where a painter always observes the distinctive features and decides the shape and orientation of the stroke, we apply techniques of image moment functions and texture analysis, from which features are extracted and used to guide the stroke generation. Techniques are also developed for dynamic determination of cropped image size and edge enhancement. The main features in the source image are well preserved.
Keywords :
feature extraction; image enhancement; image reconstruction; image representation; image texture; rendering (computer graphics); cropped image size; edge enhancement; feature extraction; image moment functions; image rendering; nonphotorealistic rendering; rectangular brush strokes; stroke-based painting; texture analysis; Digital images; High definition video; Histograms; Image analysis; Image color analysis; Image edge detection; Image sequence analysis; Image texture analysis; Painting; Rendering (computer graphics);
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Image Processing. 2002. Proceedings. 2002 International Conference on
ISSN :
1522-4880
Print_ISBN :
0-7803-7622-6
Type :
conf
DOI :
10.1109/ICIP.2002.1038109
Filename :
1038109
Link To Document :
بازگشت