Graph-FCN for Image Semantic Segmentation

By Yi Lu, Yaran Chen, Dongbin Zhao, Jianxin Chen. 2019

154 citations recorded by the source corpus.

Semantic segmentation with deep learning has achieved great progress in classifying the pixels in the image. However, the local location information is usually ignored in the high-level feature extraction by the deep learning, which is important for image semantic segmentation. To avoid this problem, we propose a graph model initialized by a fully convolutional network (FCN) named Graph-FCN for image semantic segmentation. Firstly, the image grid data is extended to graph structure data by a convolutional network, which transforms the semantic segmentation problem into a graph node classification problem. Then we apply graph convolutional network to solve this graph node classification problem. As far as we know, it is the first time that we apply the graph convolutional network in image semantic segmentation. Our method achieves competitive performance in mean intersection over union (mIOU) on the VOC dataset (about 1.34% improvement), compared to the original FCN model.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Wenhai Wang
2. Deng-Ping Fan
3. Tong Lu
4. Yirui Wu
5. Ling Shao
6. Huan Fu
7. Guangchen Shi
8. Jinpeng Li

In Computer Vision and Pattern Recognition

1. Ling Shao
2. Deng-Ping Fan
3. Huan Fu
4. Wenhai Wang
5. Tong Lu
6. Ming-Ming Cheng
7. Luc Van Gool
8. Jianbing Shen

Citations in the corpus

• Incremental Few-Shot Semantic Segmentation via Embedding Adaptive-Update and Hyper-class Representation (2022, methodology): “Using topological structure of graphs, GNN-based methods [30, 32, 42] transform task of image segmentation into the classification task of graph nodes.”
• Polyp-PVT: Polyp Segmentation with Pyramid Vision Transformers (2021, methodology): “We introduce the non-local [80], [81] operation under graph convolution domain [82] to implement our similarity aggregation module (SAM).”