A Deeper Look into Second-Order Feature Aggregation for LiDAR Place Recognition

2024-09-24

AI Summary

To address the storage and computational overhead of high-dimensional global point cloud descriptors in LiDAR-based place recognition, this paper proposes Channel-Partitioned Second-order aggregation (CPS), a lightweight second-order feature aggregation method. CPS innovatively integrates channel grouping with local covariance modeling—preserving all channel information without loss—while requiring only four learnable parameters. It comprises three components: channel-partitioned second-order aggregation, lightweight covariance modeling, and compact descriptor generation, and supports cross-dataset generalization training. Evaluated on four major benchmarks—Oxford RobotCar, In-house, MulRan, and WildPlaces—CPS achieves new state-of-the-art performance, outperforming full-covariance and random-projection baselines. The results demonstrate its superior efficiency and robustness under resource-constrained conditions, while retaining essential second-order statistical characteristics critical for discriminative representation learning.

Technology Category

• Computer Vision: Learning & Optimization for CV
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Multimodal Learning

Application Category

• Security and Privacy: Large-scale security measurements
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Efficient LiDAR Place Recognition (LPR) compresses dense pointwise features into compact global descriptors. While first-order aggregators such as GeM and NetVLAD are widely used, they overlook inter-feature correlations that second-order aggregation naturally captures. Full covariance, a common second-order aggregator, is high in dimensionality; as a result, practitioners often insert a learned projection or employ random sketches -- both of which either sacrifice information or increase parameter count. However, no prior work has systematically investigated how first- and second-order aggregation perform under constrained feature and compute budgets. In this paper, we first demonstrate that second-order aggregation retains its superiority for LPR even when channels are pruned and backbone parameters are reduced. Building on this insight, we propose Channel Partition-based Second-order Local Feature Aggregation (CPS): a drop-in, partition-based second-order aggregation module that preserves all channels while producing an order-of-magnitude smaller descriptor. CPS matches or exceeds the performance of full covariance and outperforms random projection variants, delivering new state-of-the-art results with only four additional learnable parameters across four large-scale benchmarks: Oxford RobotCar, In-house, MulRan, and WildPlaces.

Problem

• Investigates second-order feature aggregation for LiDAR place recognition
• Compares first- and second-order aggregation under constrained budgets
• Proposes CPS for efficient second-order aggregation with minimal parameters

Innovation

• Second-order aggregation for LiDAR place recognition
• Channel Partition-based Second-order Local Feature Aggregation
• Compact descriptors with minimal additional parameters