Learning Graph Convolutional Network for Blind Mesh Visual Quality Assessment

Ilyass Abouelaziz,Aladine Chetouani,Longin Jan Latecki,Mohammed El Hassouni,Hocine Cherifi

doi:10.1109/access.2021.3094663

Abstract

This paper proposes a new method for blind mesh visual quality assessment (MVQA) based on a graph convolutional network. For that, we address the node classification problem to predict the perceived visual quality. First, two matrices representing the 3D mesh are considered: a graph adjacency matrix and a feature matrix. Both matrices are used as input to a shallow graph convolutional network. The network consists of two convolutional layers followed by a max-pooling layer to provide the final feature representation. With this structure, the Softmax classifier predicts the quality score category without the reference mesh's availability. Experiments are conducted on four publicly available databases constructed explicitly for the mesh quality assessment task. We investigate several perceptual and visual features to select the most effective combination. Comparisons with the state-of-the-art alternative methods show the effectiveness of the proposed framework.

Highlights

ECENTLY, the issue of perceptual 3D mesh visual quality assessment (MVQA) has become an essential field of study since 3D models are widely used in a diversity of applications
We propose here a model-based method 117 using a shallow Graph Convolutional Network (GCN) to process the 3D model itself directly
Several sets used as input to the GCN have been tested

Summary

Introduction

ECENTLY, the issue of perceptual 3D MVQA has become an essential field of study since 3D models are widely used in a diversity of applications. One can classify objective methods according to the availability of the reference object. In NoReference (NR) or Blind methods, no information about the reference is available [8]–[12]. We denote n and m the number of vertices and edges, respectively. Adjacency matrix: The adjacency matrix A of size n × n representing a graph is defined as Ai,j = 1 if the vertices vi and vj are connected by an edge (i.e. adjacent vertices), otherwise Ai,j = 0. Each node and edge may have attribute values which are considered as features of the graph. The term attribute value is used instead of label to make the distinction with the concept of labeling in graph-theory. A walk is a sequence of nodes in a graph, in which consecutive nodes are connected by an edge. N1 (v) is the 1-neighborhood of a node, that is, all nodes that are adjacent to v

Methods

Results

Conclusion