Journal
VISUAL COMPUTER
Volume 36, Issue 7, Pages 1307-1324Publisher
SPRINGER
DOI: 10.1007/s00371-019-01726-2
Keywords
Deep neural networks; Style transfer; Soft mask; Semantic segmentation
Categories
Funding
- National Natural Science Foundation of China [61503128]
- Science and Technology Plan Project of Hunan Province [2016TP1020]
- Scientific Research Fund of Hunan Provincial Education Department [16C0226, 18A333]
- Hengyang guided science and technology projects and Application-oriented Special Disciplines [Hengkefa [2018]60-31]
- Double FirstClass University Project of Hunan Province [Xiangjiaotong [2018]469]
- Hunan Province Special Funds of Central Government for Guiding Local Science and Technology Development [2018CT5001]
- Subject Group Construction Project of Hengyang Normal University [18XKQ02, 61733004]
Ask authors/readers for more resources
This paper presents an automatic image synthesis method to transfer the style of an example image to a content image. When standard neural style transfer approaches are used, the textures and colours in different semantic regions of the style image are often applied inappropriately to the content image, ignoring its semantic layout and ruining the transfer result. In order to reduce or avoid such effects, we propose a novel method based on automatically segmenting the objects and extracting their soft semantic masks from the style and content images, in order to preserve the structure of the content image while having the style transferred. Each soft mask of the style image represents a specific part of the style image, corresponding to the soft mask of the content image with the same semantics. Both the soft masks and source images are provided as multichannel input to an augmented deep CNN framework for style transfer which incorporates a generative Markov random field model. The results on various images show that our method outperforms the most recent techniques.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available