SelFLoc: Selective Feature Fusion for Large-scale Point Cloud-based Place Recognition

TL;DR

SelFLoc introduces a novel architecture combining asymmetric convolution and selective feature fusion to improve large-scale point cloud-based place recognition, achieving state-of-the-art results on multiple benchmarks.

Contribution

The paper proposes SACB and SFFB modules for enhanced feature representation and fusion in point cloud place recognition, advancing the state-of-the-art performance.

Findings

Achieves 1.6% higher recall@1 than previous methods.

Demonstrates effectiveness of asymmetric convolution in point cloud features.

Outperforms existing methods on Oxford and in-house benchmarks.

Abstract

Point cloud-based place recognition is crucial for mobile robots and autonomous vehicles, especially when the global positioning sensor is not accessible. LiDAR points are scattered on the surface of objects and buildings, which have strong shape priors along different axes. To enhance message passing along particular axes, Stacked Asymmetric Convolution Block (SACB) is designed, which is one of the main contributions in this paper. Comprehensive experiments demonstrate that asymmetric convolution and its corresponding strategies employed by SACB can contribute to the more effective representation of point cloud feature. On this basis, Selective Feature Fusion Block (SFFB), which is formed by stacking point- and channel-wise gating layers in a predefined sequence, is proposed to selectively boost salient local features in certain key regions, as well as to align the features before fusion phase. SACBs and SFFBs are combined to construct a robust and accurate architecture for point cloud-based place recognition, which is termed SelFLoc. Comparative experimental results show that SelFLoc achieves the state-of-the-art (SOTA) performance on the Oxford and other three in-house benchmarks with an improvement of 1.6 absolute percentages on mean average recall@1.