Self-Supervised Scene Flow Estimation with 4-D Automotive Radar

TL;DR

This paper introduces a self-supervised method for estimating scene flow from 4-D automotive radar data, overcoming challenges of data sparsity and noise, and demonstrating robustness in real-world autonomous driving scenarios.

Contribution

It presents a novel self-supervised learning framework and specialized loss functions tailored for radar data, enabling effective scene flow estimation without annotated datasets.

Findings

Robust scene flow estimation from radar data in real-world conditions

Improved motion segmentation using radar scene flow

Effective handling of sparse and noisy radar data

Abstract

Scene flow allows autonomous vehicles to reason about the arbitrary motion of multiple independent objects which is the key to long-term mobile autonomy. While estimating the scene flow from LiDAR has progressed recently, it remains largely unknown how to estimate the scene flow from a 4-D radar - an increasingly popular automotive sensor for its robustness against adverse weather and lighting conditions. Compared with the LiDAR point clouds, radar data are drastically sparser, noisier and in much lower resolution. Annotated datasets for radar scene flow are also in absence and costly to acquire in the real world. These factors jointly pose the radar scene flow estimation as a challenging problem. This work aims to address the above challenges and estimate scene flow from 4-D radar point clouds by leveraging self-supervised learning. A robust scene flow estimation architecture and three novel losses are bespoken designed to cope with intractable radar data. Real-world experimental results validate that our method is able to robustly estimate the radar scene flow in the wild and effectively supports the downstream task of motion segmentation.