TY - JOUR
T1 - Map art style transfer with multi-stage framework
AU - Shih, Chiao Yin
AU - Chen, Ya Hsuan
AU - Lee, Tong Yee
N1 - Funding Information:
In this paper, we have presented a multi-stage framework for map art style transfer, which can be used to reduce the noise map art images and produce a better harmonization in such these results. The proposed framework utilizes a reference style image and a content image to produce an initial map art. A refinement strategy is then employed to generate better map art images. We demonstrate the effectiveness of our framework by an objective evaluation with those of other works. Experimental results prove that the proposed scheme performs well among other methods. We finally show a bad example and in the future, we plan to explore a solution to cope with this problem. Besides, it will be interesting to first remove the original photo from map art to re-generate the map and then combine a new photo with this re-generated map. It will be more challenging how to recover the original map successfully and create a new map art with this regenerated map in our future research. This work was supported in part by the Ministry of Science and Technology under 108-2221-E-006-038-MY3, and 107-2221-E-006-196-MY3, Taiwan. We would like to thank Ms. Hanh Le, our senior PhD student in our Laboratory, for insight assistance with methodology, source code that greatly improved the manuscript.
Publisher Copyright:
© 2020, Springer Science+Business Media, LLC, part of Springer Nature.
PY - 2021/1
Y1 - 2021/1
N2 - We propose a multi-stage framework to create the stylized map art images. Existing techniques are successful in transferring style in photos. Yet, the noise in results and the harmonization in the generated art images still need to be investigated. We address these issues with a proposed algorithm that defines a good portrait for map art application in the initial round. A refinement strategy is then applied to produce the final map arts that meet the aforementioned expectations. Beside our plausible results, the objective evaluation presented in this paper shows that our proposed method can interactively achieve better and appealing map art results in the comparison with those of other works. In addition, our method can also create ocean or landscape stylized paintings using our map art collage.
AB - We propose a multi-stage framework to create the stylized map art images. Existing techniques are successful in transferring style in photos. Yet, the noise in results and the harmonization in the generated art images still need to be investigated. We address these issues with a proposed algorithm that defines a good portrait for map art application in the initial round. A refinement strategy is then applied to produce the final map arts that meet the aforementioned expectations. Beside our plausible results, the objective evaluation presented in this paper shows that our proposed method can interactively achieve better and appealing map art results in the comparison with those of other works. In addition, our method can also create ocean or landscape stylized paintings using our map art collage.
UR - http://www.scopus.com/inward/record.url?scp=85091688563&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85091688563&partnerID=8YFLogxK
U2 - 10.1007/s11042-020-09788-4
DO - 10.1007/s11042-020-09788-4
M3 - Article
AN - SCOPUS:85091688563
SN - 1380-7501
VL - 80
SP - 4279
EP - 4293
JO - Multimedia Tools and Applications
JF - Multimedia Tools and Applications
IS - 3
ER -