CloudWalker: Random walks for 3D point cloud shape analysis

TL;DR

CloudWalker introduces a novel approach for 3D shape analysis by leveraging random walks on point clouds, enabling effective learning despite their irregular structure, and achieves state-of-the-art results in classification and retrieval tasks.

Contribution

It presents a new method that uses random walks to impose structure on point clouds for improved 3D shape learning, differing from prior grid or mesh-based approaches.

Findings

Achieves state-of-the-art results in 3D shape classification.

Effective in 3D shape retrieval tasks.

Demonstrates robustness on irregular point cloud data.

Abstract

Point clouds are gaining prominence as a method for representing 3D shapes, but their irregular structure poses a challenge for deep learning methods. In this paper we propose CloudWalker, a novel method for learning 3D shapes using random walks. Previous works attempt to adapt Convolutional Neural Networks (CNNs) or impose a grid or mesh structure to 3D point clouds. This work presents a different approach for representing and learning the shape from a given point set. The key idea is to impose structure on the point set by multiple random walks through the cloud for exploring different regions of the 3D object. Then we learn a per-point and per-walk representation and aggregate multiple walk predictions at inference. Our approach achieves state-of-the-art results for two 3D shape analysis tasks: classification and retrieval.