HRGE-Net: Hierarchical Relational Graph Embedding Network for Multi-view 3D Shape Recognition

TL;DR

HRGE-Net introduces a hierarchical relational graph embedding network that effectively aggregates multi-view features for 3D shape recognition, achieving state-of-the-art results in classification and retrieval tasks.

Contribution

The paper proposes a novel hierarchical relational graph embedding network that models view relations for improved multi-view 3D shape recognition.

Findings

Achieves state-of-the-art performance on benchmark datasets.

Effectively models pairwise and neighboring view relations.

Improves 3D shape classification and retrieval accuracy.

Abstract

View-based approach that recognizes 3D shape through its projected 2D images achieved state-of-the-art performance for 3D shape recognition. One essential challenge for view-based approach is how to aggregate the multi-view features extracted from 2D images to be a global 3D shape descriptor. In this work, we propose a novel feature aggregation network by fully investigating the relations among views. We construct a relational graph with multi-view images as nodes, and design relational graph embedding by modeling pairwise and neighboring relations among views. By gradually coarsening the graph, we build a hierarchical relational graph embedding network (HRGE-Net) to aggregate the multi-view features to be a global shape descriptor. Extensive experiments show that HRGE-Net achieves stateof-the-art performance for 3D shape classification and retrieval on benchmark datasets.