Unsupervised 4D LiDAR Moving Object Segmentation in Stationary Settings with Multivariate Occupancy Time Series

TL;DR

This paper introduces an unsupervised method for segmenting moving objects in 4D LiDAR data from stationary sensors by modeling occupancy changes as multivariate time series and clustering features, eliminating the need for ground truth annotations.

Contribution

It proposes a novel 4D LiDAR representation using multivariate occupancy time series and a self-supervised neural network for unsupervised moving object segmentation.

Findings

Achieves comparable performance to supervised methods on KITTI dataset

Introduces a new time series clustering approach for LiDAR data

Eliminates dependence on annotated ground truth for MOS

Abstract

In this work, we address the problem of unsupervised moving object segmentation (MOS) in 4D LiDAR data recorded from a stationary sensor, where no ground truth annotations are involved. Deep learning-based state-of-the-art methods for LiDAR MOS strongly depend on annotated ground truth data, which is expensive to obtain and scarce in existence. To close this gap in the stationary setting, we propose a novel 4D LiDAR representation based on multivariate time series that relaxes the problem of unsupervised MOS to a time series clustering problem. More specifically, we propose modeling the change in occupancy of a voxel by a multivariate occupancy time series (MOTS), which captures spatio-temporal occupancy changes on the voxel level and its surrounding neighborhood. To perform unsupervised MOS, we train a neural network in a self-supervised manner to encode MOTS into voxel-level feature representations, which can be partitioned by a clustering algorithm into moving or stationary. Experiments on stationary scenes from the Raw KITTI dataset show that our fully unsupervised approach achieves performance that is comparable to that of supervised state-of-the-art approaches.