Abstract | ||
---|---|---|
This paper proposes a new method of transferring style between images by considering scene matching between the source image and the target image. Artists often employ different colors and brushwork for individual subjects. Likewise, the connections between various subjects in a work also affect the colors and brushwork used. Our method begins with input images, searches an example database for paintings with scenes similar to that in the input image, and transfers the color and brushwork of the paintings to the corresponding target images to generate painterly images that reflect specific styles. Our method applies a GIST approach to the process of searching for paintings with similar scenes before performing style transfers. The spatial correspondence between the source image and the target image is also used to ensure close correlation between various elements in order to reproduce styles faithfully. |
Year | Venue | Field |
---|---|---|
2017 | Trans. Computational Science | Computer vision,Scene matching,Computer science,Painting,Artificial intelligence |
DocType | Volume | Citations |
Journal | 30 | 0 |
PageRank | References | Authors |
0.34 | 8 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Masahiro Toyoura | 1 | 64 | 19.34 |
Noriyuki Abe | 2 | 1 | 0.70 |
Xiaoyang Mao | 3 | 351 | 58.66 |