¡@

In this paper, we present a non-photorealistic rendering framework to emulate cubist style paintings for 3D polygonal models. This research is inspired from the Picasso¡¦s paintings, i.e., the paintings in the early stage of cubism, which compose the object¡¦s elements captured from different viewpoints in a painting. To achieve this aim, our system strives to render all the interesting features of the given model on a 2D image plane without affecting the model¡¦s appearance. Specifically, in the preprocessing we emphasize the characteristic features of the model using the proposed global shape enhancement approach. Then, we determine a set of good views by analyzing the enhanced model, and duplicate a copy for each view. The duplicated models are composed and rendered together to expose the geometric features from different viewpoints. Since several models are rendered on one image, we occlude the repeated regions by finding the intersection of models. The models are enforced to intersect at the compatible and low-salient regions by an optimization process, making the composition visually pleasing. Finally, the composed model is rendered by a non-photorealistic rendering technique to generate a cubist-like painting.

Keywords: non-photorealistic rendering, cubism, feature enhancement, saliency distribution, model composition

Retrieve PDF document (201111_06.pdf)

Received June 22, 2010; revised July 1, 2010; accepted August 26, 2010.
Communicated by Tyng-Luh Liu.
^* This work was supported in part by the Landmark Program of the NCKU Top University Project (contract No. C0038), the National Science Council (No. NSC-98-2221-E-006-179), Taiwan. The authors also gratefully acknowledge the helpful comments and suggestions of Dr. Tong-Yee Lee and the reviewers.