Anzeige
Erschienen in:
2018 | OriginalPaper | Buchkapitel
Image Stylization for Thread Art via Color Quantization and Sparse Modeling
verfasst von : Kewei Yang, Zhengxing Sun, Shuang Wang, Hui-Hsia Chen
Erschienen in: Advances in Multimedia Information Processing – PCM 2017
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Abstract
We present an image stylization method to simulate a graceful Chinese art—Random-needle Embroidery designated as Intangible Cultural Heritage. We first develop an effective way to simulate a single thread, and define rendering primitive called stitch which is a collection of threads arranged in a certain pattern. Then, we segment the input image and partition each region into non-uniform sub-regions, from which a color quantization method is proposed to select colors used in the rendering process for each sub-region from a specific color library. During runtime, new stitches can be synthesized for each sub-region via sparse modeling based on the pre-defined stitches. Smoothness constraints are added to this process to avoid local distortions. Finally, rendering image is generated by placing stitches with different attributes on the canvas. Experiments show that our method can perform fine images with the style of random-needle.