Revisiting Radar Perception With Spectral Point Clouds

arXiv cs.CV / 4/10/2026

💬 OpinionSignals & Early TrendsIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper argues that radar perception models trained on dense range-Doppler spectra do not necessarily transfer better than sparse point-cloud inputs due to large sensor/configuration differences.
It proposes a “spectral point cloud” paradigm in which point clouds act as sparse, compressed representations of radar spectra to enable a more unified and sensor-robust input format.
The authors build an experimental framework that compares spectral point cloud models at different densities against a dense range-Doppler benchmark to identify the point-cloud density thresholds that match or reach benchmark performance.
Two spectral enrichment approaches are tested, and the results show point-cloud models can match dense range-Doppler performance and even surpass it when enriched with additional target-relevant spectral information.
The work positions spectral point clouds as promising building blocks toward future radar foundation models with unified representations.

Abstract

Radar perception models are trained with different inputs, from range-Doppler spectra to sparse point clouds. Dense spectra are assumed to outperform sparse point clouds, yet they can vary considerably across sensors and configurations, which hinders transfer. In this paper, we provide alternatives for incorporating spectral information into radar point clouds and show that, point clouds need not underperform compared to spectra. We introduce the spectral point cloud paradigm, where point clouds are treated as sparse, compressed representations of the radar spectra, and argue that, when enriched with spectral information, they serve as strong candidates for a unified input representation that is more robust against sensor-specific differences. We develop an experimental framework that compares spectral point cloud (PC) models at varying densities against a dense range-Doppler (RD) benchmark, and report the density levels where the PC configurations meet the performance of the RD benchmark. Furthermore, we experiment with two basic spectral enrichment approaches, that inject additional target-relevant information into the point clouds. Contrary to the common belief that the dense RD approach is superior, we show that point clouds can do just as well, and can surpass the RD benchmark when enrichment is applied. Spectral point clouds can therefore serve as strong candidates for unified radar perception, paving the way for future radar foundation models.