MinkUNeXt: Point Cloud-based Large-scale Place Recognition using 3D Sparse Convolutions

TL;DR

MinkUNeXt introduces a novel 3D sparse convolution-based architecture for large-scale place recognition from point clouds, outperforming current state-of-the-art methods without using complex Transformer components.

Contribution

The paper proposes MinkUNeXt, a new architecture utilizing simple 3D sparse convolutions and a U-Net structure for effective place recognition from point clouds.

Findings

Outperforms existing state-of-the-art methods on Oxford RobotCar dataset.

Effective feature extraction at multiple scales with a U-Net encoder-decoder.

Achieves superior accuracy using only 3D sparse convolutions without Transformers.

Abstract

This paper presents MinkUNeXt, an effective and efficient architecture for place-recognition from point clouds entirely based on the new 3D MinkNeXt Block, a residual block composed of 3D sparse convolutions that follows the philosophy established by recent Transformers but purely using simple 3D convolutions. Feature extraction is performed at different scales by a U-Net encoder-decoder network and the feature aggregation of those features into a single descriptor is carried out by a Generalized Mean Pooling (GeM). The proposed architecture demonstrates that it is possible to surpass the current state-of-the-art by only relying on conventional 3D sparse convolutions without making use of more complex and sophisticated proposals such as Transformers, Attention-Layers or Deformable Convolutions. A thorough assessment of the proposal has been carried out using the Oxford RobotCar and the In-house datasets. As a result, MinkUNeXt proves to outperform other methods in the state-of-the-art.