Séminaires des doctorants - Patty Coupeau13h30 | POLYTECH ANGERS |Salle 10 | 62, avenue Notre-Dame du Lac | ANGERS

Le 12 mai 2022

Subject : On the relevance of edge-conditioned convolution for GNN-based semantic image segmentation using spatial relationships

Abstract

This presentation addresses the fundamental task of se\u0002mantic image segmentation by exploiting structural information (spatial relationships between image regions). To perform such task, we propose to combine a deep neural network (CNN) with inexact ”many-to-one-or-none” graph matching where graphs encode efficiently class probabilities and structural information related to regions segmented by the CNN. In order to achieve node classification, a basic 2-layer graph neural network (GNN) based on the edge-conditioned convolution operator (ECConv), managing both node and edge attributes, is considered. Preliminary experiments are performed on both a synthetic dataset and a public dataset of face images (FASSEG). Our approach is shown to be resilient to small training datasets that often limit the performance of deep learning thanks to a preprocessing task of graph coarsening. Results show that the proposal reaches a perfect accuracy on synthetic dataset and improves performance of the CNN by 6% (bounding box dice index) on FASSEG. Moreover, it enhances by 27% the initial Hausdorff distance (i.e. with CNN only) using the entire training dataset and by 41% with only 75% of training samples."

Index Terms

image segmentation, structural information, in\u0002exact graph matching, graph neural network, edge-conditioned convolution

Download the slides of the presentation