Title: DCNet: diffusion convolutional networks for semantic image segmentation

Authors: Lan Yang; Zhixiong Jiang; Hongbo Zhou; Jun Guo

Addresses: School of Software, Quanzhou University of Information Engineering, Quanzhou, Fujian, China ' School of Business, Shanghai Dianji University, Shanghai, China ' School of Software, Quanzhou University of Information Engineering, Quanzhou, Fujian, China ' School of Software, Quanzhou University of Information Engineering, Quanzhou, Fujian, China

Abstract: Semantic image segmentation makes a pixel-level classification play an essential role in scene understanding. Recently, most approaches exploit deep learning neural networks, especially convolutional neural networks (CNNs), to tackle the image segmentation challenge. Common issues of these CNN-based methods are the loss of spatial features during learning representations and the limited capacity for capturing contextual information in a large receptive field. This paper proposes a diffusion convolutional network (DCNet) to combine the CNN and graph convolutional neural network (GCNN) for semantic image segmentation. In the proposed model, diffusion convolution is formulated as a graph convolutional layer to aggregate structural and contextual information without losing spatial features. The final segmentation results on the PASCAL VOC 2012 and Cityscapes datasets show better performance than baseline approaches and can be competitive with state-of-the-art methods.

Keywords: deep learning; semantic image segmentation; graph convolutional neural networks; GCNNs.

DOI: 10.1504/IJES.2021.116135

International Journal of Embedded Systems, 2021 Vol.14 No.3, pp.300 - 311

Received: 08 Jan 2021
Accepted: 11 Jan 2021
Published online: 12 Jul 2021 *