Radar Instance Transformer: Reliable Moving Instance Segmentation in Sparse Radar Point Clouds

TL;DR

This paper introduces Radar Instance Transformer, a novel method for moving instance segmentation in sparse radar point clouds, leveraging temporal data and a full-resolution backbone to improve accuracy and reliability in dynamic environments.

Contribution

It proposes a new transformer-based approach that enriches radar scans with temporal info and uses a full-resolution backbone, advancing moving instance segmentation in radar data.

Findings

Outperforms existing methods on RadarScenes dataset

Provides reliable, class-agnostic instance segmentation

Enhances scene interpretation in diverse environments

Abstract

The perception of moving objects is crucial for autonomous robots performing collision avoidance in dynamic environments. LiDARs and cameras tremendously enhance scene interpretation but do not provide direct motion information and face limitations under adverse weather. Radar sensors overcome these limitations and provide Doppler velocities, delivering direct information on dynamic objects. In this paper, we address the problem of moving instance segmentation in radar point clouds to enhance scene interpretation for safety-critical tasks. Our Radar Instance Transformer enriches the current radar scan with temporal information without passing aggregated scans through a neural network. We propose a full-resolution backbone to prevent information loss in sparse point cloud processing. Our instance transformer head incorporates essential information to enhance segmentation but also enables reliable, class-agnostic instance assignments. In sum, our approach shows superior performance on the new moving instance segmentation benchmarks, including diverse environments, and provides model-agnostic modules to enhance scene interpretation. The benchmark is based on the RadarScenes dataset and will be made available upon acceptance.