PREMA: Part-based REcurrent Multi-view Aggregation Network for 3D Shape Retrieval

TL;DR

PREMA is a novel network that improves 3D shape retrieval by focusing on discriminant parts across multiple views, overcoming view defects and enhancing shape representation robustness.

Contribution

It introduces the Part-based Recurrent Multi-view Aggregation network (PREMA) with a Regional Attention Unit to locate and utilize discriminant parts for better 3D shape retrieval.

Findings

PREMA effectively handles view defects like occlusions and background clutter.

The network improves shape discrimination by focusing on consistent discriminant parts.

Experimental results show enhanced retrieval accuracy over existing methods.

Abstract

We propose the Part-based Recurrent Multi-view Aggregation network(PREMA) to eliminate the detrimental effects of the practical view defects, such as insufficient view numbers, occlusions or background clutters, and also enhance the discriminative ability of shape representations. Inspired by the fact that human recognize an object mainly by its discriminant parts, we define the multi-view coherent part(MCP), a discriminant part reoccurring in different views. Our PREMA can reliably locate and effectively utilize MCPs to build robust shape representations. Comprehensively, we design a novel Regional Attention Unit(RAU) in PREMA to compute the confidence map for each view, and extract MCPs by applying those maps to view features. PREMA accentuates MCPs via correlating features of different views, and aggregates the part-aware features for shape representation.