Non-Photorealistic Rendering for Aesthetic Virtual Environments
Author(s): Chung-Ming Wang, Jiunn-Shyan Lee.
Article: Journal of Information Science and Engineering, Vol. 20, No. 5, pp. 923--948, September,
2004.
[BibTeX] 
Abstract:
The line integral convolution (LIC) method, originally developed for imaging vector fields in scientific visualization, has the potential to produce images with directional characteristics. In this paper, we present four techniques to explore the use of LIC in generating images in the style of the Impressionists. In particular, we develop an Impressionist Line Integral Convolution algorithm (ILIC) to generate images with impressionist styles. This algorithm takes advantage of directional information given by a photographic image, incorporates a shading technique to blend cool and warm colors into the image, and finally applies the revised LIC method to imitate artists, paintings done in impressionist styles. Furthermore, we propose a color fidelity technique, which takes advantage of the cool-to-warm scheme, initiating artists' conventional painting to enhance visual depth perception. We also present an information preservation technique, which quantifies image details to control the convolution length, making it possible to preserve subtle information during the convolution process. Finally, we demonstrate a top-down sampling technique, where a series of artistic mip-maps is generated to construct aesthetic virtual environments. These maps provide constant strokes with directional cues, thus achieving frame-to-frame coherence in an interactive walkthrough system. Both the silhouette drawing and tour into the picture (TIP) approaches are employed to enhance the user's immersion in a virtual world. The experimental results demonstrate the merits of our techniques in generating images in impressionist styles and constructing an interactive walkthrough system that provides an immersion experience rendered in painting styles.