# GeoAuxNet: Towards Universal 3D Representation Learning for Multi-sensor Point Clouds

Source: [GeoAuxNet: Towards Universal 3D Representation Learning for Multi-sensor Point Clouds](<https://www.sota2.com/research/paper/68fa72ad1e209131cea57cf0>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-01-04T04:42:42.961Z

Authors: Shengjun Zhang, Xin Fei, Yueqi Duan

Published: 2024-03-28T08:34:04.000Z

## Abstract

Point clouds captured by different sensors such as RGB-D cameras and LiDAR possess non-negligible domain gaps. Most existing methods design different network architectures and train separately on point clouds from various sensors. Typically, point-based methods achieve outstanding performances on even-distributed dense point clouds from RGB-D cameras, while voxel-based methods are more efficient for large-range sparse LiDAR point clouds. In this paper, we propose geometry-to-voxel auxiliary learning to enable voxel representations to access point-level geometric information, which supports better generalisation of the voxel-based backbone with additional interpretations of multi-sensor point clouds. Specifically, we construct hierarchical geometry pools generated by a voxel-guided dynamic point network, which efficiently provide auxiliary fine-grained geometric information adapted to different stages of voxel features. We conduct experiments on joint multi-sensor datasets to demonstrate the effectiveness of GeoAuxNet. Enjoying elaborate geometric information, our method outperforms other models collectively trained on multi-sensor datasets, and achieve competitive results with the-state-of-art experts on each single dataset.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [3D Semantic Segmentation on S3DIS Area 5 (test)](<https://www.sota2.com/research/sota/3d-semantic-segmentation-on-s3dis-area-5-test.md>): 3D Semantic Segmentation; S3DIS Area 5 (test); mIoU (%): 69.5.

- [Semantic Segmentation on ScanNet (val)](<https://www.sota2.com/research/sota/semantic-segmentation-on-scannet-val.md>): Semantic segmentation; ScanNet (val); mIoU: 71.3.

- [Semantic Segmentation on SemanticKITTI (val)](<https://www.sota2.com/research/sota/semantic-segmentation-on-semantickitti-val.md>): Semantic segmentation; SemanticKITTI (val); mIoU: 63.8.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2403.19220>)

[PDF](<https://arxiv.org/pdf/2403.19220.pdf>)

[Implementation](<https://github.com/zhangshengjun2019/GeoAuxNet>)