Boundary-aware Graph Reasoning for Semantic Segmentation
This work addresses semantic segmentation accuracy, particularly for boundary regions, which is an incremental improvement for computer vision applications.
The paper tackles the problem of semantic segmentation by proposing a Boundary-aware Graph Reasoning (BGR) module to improve long-range contextual feature learning, focusing on boundary regions where segmentation errors commonly occur, and demonstrates effectiveness through experiments on three benchmarks.
In this paper, we propose a Boundary-aware Graph Reasoning (BGR) module to learn long-range contextual features for semantic segmentation. Rather than directly construct the graph based on the backbone features, our BGR module explores a reasonable way to combine segmentation erroneous regions with the graph construction scenario. Motivated by the fact that most hard-to-segment pixels broadly distribute on boundary regions, our BGR module uses the boundary score map as prior knowledge to intensify the graph node connections and thereby guide the graph reasoning focus on boundary regions. In addition, we employ an efficient graph convolution implementation to reduce the computational cost, which benefits the integration of our BGR module into current segmentation backbones. Extensive experiments on three challenging segmentation benchmarks demonstrate the effectiveness of our proposed BGR module for semantic segmentation.