Paper
14 November 2023 Semantic image synthesis with vision graph neural network
Zhihong Xu, Xiaozhao Qian, Peng Wang
Author Affiliations +
Proceedings Volume 12934, Third International Conference on Computer Graphics, Image, and Virtualization (ICCGIV 2023); 129340L (2023) https://doi.org/10.1117/12.3008021
Event: 2023 3rd International Conference on Computer Graphics, Image and Virtualization (ICCGIV 2023), 2023, Nanjing, China
Abstract
Semantic image synthesis aims to generate realistic images from semantic label maps. Current generative adversarial network models still struggle with understanding visual information with irregular topological structures because they rely on convolutional neural networks, which interpret semantic label map as grid structures. In this work, we propose a Vision GNN Generative Adversarial Network (VG-GAN). In our redesigned generator, the input semantic label map is embedded in patches and then converted into a graph structure. We use graph neural networks on the constructed graph to learn the complex interrelationships in the graph structure. Experimental results show that our proposed method generates images with irregular and complex objects that appear more realistic and perform better than state-of-the-art methods.
(2023) Published by SPIE. Downloading of the abstract is permitted for personal use only.
Zhihong Xu, Xiaozhao Qian, and Peng Wang "Semantic image synthesis with vision graph neural network", Proc. SPIE 12934, Third International Conference on Computer Graphics, Image, and Virtualization (ICCGIV 2023), 129340L (14 November 2023); https://doi.org/10.1117/12.3008021
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Semantics

Neural networks

Image processing

Image segmentation

Convolution

Back to Top