SD4R: Sparse-to-Dense Learning for 3D Object Detection with 4D Radar

TL;DR

SD4R introduces a novel framework that densifies sparse 4D radar point clouds for improved 3D object detection, demonstrating state-of-the-art results on a public dataset.

Contribution

The paper presents SD4R, a new method combining a foreground point generator and logit-query encoder to densify radar point clouds and enhance detection robustness.

Findings

Achieves state-of-the-art detection performance on View-of-Delft dataset.

Effectively reduces noise and densifies sparse radar point clouds.

Demonstrates robustness in scenes with few points.

Abstract

4D radar measurements offer an affordable and weather-robust solution for 3D perception. However, the inherent sparsity and noise of radar point clouds present significant challenges for accurate 3D object detection, underscoring the need for effective and robust point clouds densification. Despite recent progress, existing densification methods often fail to address the extreme sparsity of 4D radar point clouds and exhibit limited robustness when processing scenes with a small number of points. In this paper, we propose SD4R, a novel framework that transforms sparse radar point clouds into dense representations. SD4R begins by utilizing a foreground point generator (FPG) to mitigate noise propagation and produce densified point clouds. Subsequently, a logit-query encoder (LQE) enhances conventional pillarization, resulting in robust feature representations. Through these innovations, our SD4R demonstrates strong capability in both noise reduction and foreground point densification. Extensive experiments conducted on the publicly available View-of-Delft dataset demonstrate that SD4R achieves state-of-the-art performance. Source code is available at https://github.com/lancelot0805/SD4R.