Semantic image segmentation using multi-view graph neural network

doi:10.1016/j.image.2026.117539

Semantic image segmentation using multi-view graph neural network

2026-03-16

0

PRE

AI

Jean-Baptiste Fasquel

DOI：10.1016/j.image.2026.117539

Original

Original request for help

Save

Abstract

En 中文

• MVGNN integrates multiple graph views for improved segmentation accuracy. • Early graph fusion is used with GTN, refining boundaries for better segmentation. • MVGNN with CNN outperforms CNN-based methods across multiple segmentation metrics. • Optimal graph convolution operator selection boosts model efficiency.

Keywords:

MVGNN

graph neural network

semantic image segmentation

early graph fusion

graph convolution operator

Semantic image segmentation using multi-view graph neural network

Abstract

Journal

Organization