4.4 Article

GAGCN: Generative adversarial graph convolutional network for non-homogeneous texture extension synthesis

Journal

IET IMAGE PROCESSING
Volume 17, Issue 5, Pages 1603-1614

Publisher

WILEY
DOI: 10.1049/ipr2.12741

Keywords

attention mechanism; generative adversarial networks; graph convolutional networks; non-homogeneous texture synthesis

Ask authors/readers for more resources

In non-homogeneous texture synthesis, it is important to maintain consistent overall visual characteristics when extending local patterns. Existing methods focus on local visual features and neglect the crucial relative position features for non-homogeneous texture synthesis. Therefore, modeling pixel dependence is desired to enhance synthesis performance. This paper proposes a non-homogeneous texture extended synthesis model (GAGCN) that combines generate adversarial network (GAN) and graph convolutional network (GCN) to ensure synthesis results from both local detail structure and overall structure.
In the non-homogeneous texture synthesis task, the overall visual characteristics should be consistent when extending the local patterns of the exemplar. The existing methods mainly focus on the local visual features of patterns but ignore the relative position features that are important for non-homogeneous texture synthesis. Although these methods have achieved success on homogeneous textures, they cannot perform well on non-homogeneous textures. Thus, it is desirable to model the dependence between pixels to improve the synthesis performance. To ensure synthesis results from both the local detail structure and the overall structure, this paper proposes a non-homogeneous texture extended synthesis model (GAGCN) combining the generate adversarial network (GAN) and the graph convolutional network (GCN). The GAN learns the internal distribution of image patches, which makes the synthetic image have rich local details. The GCN learns the latent dependence between pixels according to the statistical characteristics of the image. Based on this, a novel graph similarity loss is proposed. This loss describes the latent spatial differences between the sample image and the generated image, which helps the model to better capture global features. Experiments show that our method outperforms existing methods on non-homogeneous textures.

Authors

I am an author on this paper
Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.4
Not enough ratings

Secondary Ratings

Novelty
-
Significance
-
Scientific rigor
-
Rate this paper

Recommended

No Data Available
No Data Available