Name
Abstract | ||
|---|---|---|
Different image styles play a significant role in the human vision. Image rendering methods with non-photorealistic rendering based can simulate different illustrations and increase its aesthetic appeal. Despite many kinds of methods have been put forward to obtain various styles, technical subtleties and stylistic potential of the embroidery simulation are litter attention. This paper offers a detailed review of the embroidery art style simulating approach from a 2D photograph, and performs an evaluation features for these tasks. The primary novelty of this method is that the stitch features are generated through an embroidery stroke model, and stitch stoke will be merged to source image. Therefore, it avoids irregular needling embroidery, and highlights the stereoscopic effect which is not revealed in other rendering methods. Firstly, we generate noise image through gray adaptive method to guide the embroidery lines produced. After that, an improved line integral convolution technique is presented to generate stitch strokes, and scattered noise is normalizing to a certain line based on Hough transform. Next, the paper focuses on the raised strokes, which are rendered and obtained through bulging process technique in this paper. Finally, we can exploit mergence strategy based on mapping method to produce embroidery art style. To demonstrate the performance of our proposed method, this paper compares its simulating results with the real embroidery work and measure of image MSSIM is also used to evaluate the simulation quality. In all cases, the experimental results show that the proposed method can achieve embroidery style stitch visual quality and rich the aesthetic expression. |
Year | DOI | Venue |
|---|---|---|
2019 | 10.1007/s11042-018-6002-9 | Multimedia Tools Appl. |
Keywords | Field | DocType |
Non-photorealistic rendering, Miao embroidery, Bump texture, Color simulation, Image mergence | Computer vision,Pattern recognition,Adaptive method,Stereoscopy,Computer science,Hough transform,Non-photorealistic rendering,Exploit,Artificial intelligence,Novelty,Rendering (computer graphics),Line integral convolution | Journal |
Volume | Issue | ISSN |
78 | 1 | 1573-7721 |
Citations | PageRank | References |
0 | 0.34 | 20 |
Authors | ||
5 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Wenhua Qian | 1 | 2 | 6.11 |
| Dan Xu | 2 | 201 | 52.67 |
| Jinde Cao | 3 | 11399 | 733.03 |
| Zheng Guan | 4 | 0 | 2.03 |
| Yuanyuan Pu | 5 | 0 | 2.37 |