Voronoi-based Second-order Descriptor with Whitened Metric in LiDAR Place Recognition
This work addresses a domain-specific problem in robotics and autonomous systems by improving place recognition accuracy, though it is incremental as it builds on existing second-order pooling methods.
The paper tackled the problem of LiDAR place recognition by proposing a novel pooling method that integrates second-order pooling with Vorenoi cells and whitening to improve descriptor suitability for Euclidean distancing, achieving performance gains on benchmarks like Oxford Robotcar and Wild-Places.
The pooling layer plays a vital role in aggregating local descriptors into the metrizable global descriptor in the LiDAR Place Recognition (LPR). In particular, the second-order pooling is capable of capturing higher-order interactions among local descriptors. However, its existing methods in the LPR adhere to conventional implementations and post-normalization, and incur the descriptor unsuitable for Euclidean distancing. Based on the recent interpretation that associates NetVLAD with the second-order statistics, we propose to integrate second-order pooling with the inductive bias from Voronoi cells. Our novel pooling method aggregates local descriptors to form the second-order matrix and whitens the global descriptor to implicitly measure the Mahalanobis distance while conserving the cluster property from Voronoi cells, addressing its numerical instability during learning with diverse techniques. We demonstrate its performance gains through the experiments conducted on the Oxford Robotcar and Wild-Places benchmarks and analyze the numerical effect of the proposed whitening algorithm.