Fast 3D point clouds retrieval for Large-scale 3D Place Recognition

TL;DR

This paper introduces a novel method for fast large-scale 3D point cloud retrieval by adapting a transformer-based Differentiable Search Index, enabling constant-time retrieval for 3D place recognition tasks.

Contribution

It presents a new approach that leverages Vision Transformers and DSI to significantly accelerate 3D point cloud retrieval, outperforming existing methods in speed and accuracy.

Findings

Achieves constant-time retrieval in large-scale 3D point cloud datasets.

Outperforms state-of-the-art methods in place recognition accuracy.

Demonstrates significant speed improvements without sacrificing quality.

Abstract

Retrieval in 3D point clouds is a challenging task that consists in retrieving the most similar point clouds to a given query within a reference of 3D points. Current methods focus on comparing descriptors of point clouds in order to identify similar ones. Due to the complexity of this latter step, here we focus on the acceleration of the retrieval by adapting the Differentiable Search Index (DSI), a transformer-based approach initially designed for text information retrieval, for 3D point clouds retrieval. Our approach generates 1D identifiers based on the point descriptors, enabling direct retrieval in constant time. To adapt DSI to 3D data, we integrate Vision Transformers to map descriptors to these identifiers while incorporating positional and semantic encoding. The approach is evaluated for place recognition on a public benchmark comparing its retrieval capabilities against state-of-the-art methods, in terms of quality and speed of returned point clouds.