SVT-Net: Super Light-Weight Sparse Voxel Transformer for Large Scale Place Recognition

TL;DR

SVT-Net is a super lightweight 3D point cloud model that effectively captures both local and long-range features for large scale place recognition, achieving state-of-the-art accuracy with minimal model size.

Contribution

The paper introduces SVT-Net, a novel lightweight network combining Atom-based and Cluster-based Sparse Voxel Transformers for improved place recognition.

Findings

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

Maintains high speed with a model size of only 0.9M.

Simplified versions further reduce size to 0.8M and 0.4M while preserving performance.

Abstract

Point cloud-based large scale place recognition is fundamental for many applications like Simultaneous Localization and Mapping (SLAM). Although many models have been proposed and have achieved good performance by learning short-range local features, long-range contextual properties have often been neglected. Moreover, the model size has also become a bottleneck for their wide applications. To overcome these challenges, we propose a super light-weight network model termed SVT-Net for large scale place recognition. Specifically, on top of the highly efficient 3D Sparse Convolution (SP-Conv), an Atom-based Sparse Voxel Transformer (ASVT) and a Cluster-based Sparse Voxel Transformer (CSVT) are proposed to learn both short-range local features and long-range contextual features in this model. Consisting of ASVT and CSVT, SVT-Net can achieve state-of-the-art on benchmark datasets in terms of both accuracy and speed with a super-light model size (0.9M). Meanwhile, two simplified versions of SVT-Net are introduced, which also achieve state-of-the-art and further reduce the model size to 0.8M and 0.4M respectively.