Font Size: a A A

Techniques On Non-Photorealistic Rendering

Posted on:2004-05-24Degree:MasterType:Thesis
Country:ChinaCandidate:X F MiFull Text:PDF
GTID:2168360092470351Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
Based on the compare between photorealistic and non-photorealistic, the author gives a systematic survey on the objective, methods and the state of art of non-photorealistic graphics, and then presents some key techniques on non-photorealistic rendering.The author first proposes a shader based multiple-pass rendering model towards computer-generated non-photorealistic rendering system. Techniques concerning coordination between passes and illustration of textures and tone through strokes are described in detail. The proposed parametric texture shader language successfully enables third-party programmer define his own procedural texture. The proposed approach takes different input and output from traditional shading language to express texture, material, color and brightness of surface. Further more, by combining existing texture shaders, expressive ability of the proposed system got significantly enhances.As one of the filter instruction for shader language, this thesis then gives an implementation of image filter by automatic reconstruction of strokes from image based on image-moment Given a photo(or image) and the size of color sampling segment as inputs, the proposed system creates initial point stokes by computing the image-moment, then fuses these strokes into longer, more smooth strokes, and distills the strokes' parameters such as size, center coordinate, color etc. Finally, the resulting image is drawn by the hair brush model with desired parameters. A framework of this system is described, and a few watercolor painting samples are given to demonstrate this approach.In the previous works, virtual artistic medias are essential, whether for stroke visualization or for painting system. The author also proposes a virtual brush model based on droplet operation to simulate Chinese calligraphy and traditional Chinese painting in real time. Two ways of adopting the droplet model into virtual calligraphy and painting is discussed in detail. The later droplet model is more elaborated and can produce more vivid results while is slightly more time-costing. The novel feature of the proposed droplet virtual brush model successfully enables the simulation painting system to overcome the poor expressional ability of virtual brush built on particle system and avoids the complex evaluation of physical brush with solid model. The model, derived from the actual calligraphy and painting experience, and due to the simplicity of the droplet operation and its powerful expressive ability, considerably improves the performance of the simulation system and maintains painting effect comparable with real brush by supporting special Chinese brush effect such as dry brush, feng and stroke diffusion. On the other hand, the proposed droplet model helps to compute stroke area with well-defined geometry information and leads to the feasibility to retrieve the outlines of characters with well-defined geometry representation and the proper droplet operations discussed make it possible to express dry brush and stroke fork effect for the generated stoke outlines and character modeling. Detailed algorithms are also presented.Finally, the thesis gives some conclusions and also some proposals of the future development of the non-photorealist graphics research.In the proposed three aspects of the work, shader is basically designed for 3D scene rendering whereas stroke reconstruction is for 2D images. The visualization of these work ultimately depends on virtual medias, one of which ?virtual hairy brush ?is also introduced in this thesis. These researches are all basic techniques of non-photorealistic rendering and they have enriched the content of non-photorealistic graphics.
Keywords/Search Tags:Computer graphics, photorealistic, non-photorealistic, non-photorealistic rendering, non-photorealistic animation, painting system, computer vision, image recognition, shader, filter, image moment, droplet
PDF Full Text Request
Related items