AI Paper Atlas

Research map / Point clouds / Point cloud classification

Rethinking Radar 3D Perception: Representation Learning from Raw Spectra

Miao Zhang, Mohamed Abdelsamad, Bin Yang, Sherif Abdulatif, Benedikt Loesch, Marco Altmann, Abhinav Valada

NeurIPS 2026 · Point clouds › Point cloud classification · approach: Point-language and open-vocabulary 3D · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Challenges the view that raw radar spectra are too noisy for direct learning, arguing the missing piece is a learned representation rather than the raw data itself.

Problem

Radar 3D perception typically relies on handcrafted reductions, sparse point clouds, or compact Doppler descriptors because raw radar spectra are viewed as too cluttered and noisy for direct representation learning.

Approach

Argues the main limitation is the absence of a learned representation over raw spectra rather than the spectra themselves, and proposes learning representations directly from raw radar spectra for 3D perception. Full method details are not available from the abstract.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace point cloud with dense RAE/RAE-d spectra under supervised trainingPTv3 · K-Radar official splitmAP AP_BEV: 49.0 -> 47.4 (RAE), 45.8 (RAE-d)Hurt
add SPMAE/L-SPMAE self-supervised pretraining on spectraPTv3 · K-Radar official splitmAP AP_BEV: 48.9 -> 55.9 (RAD)Improved
add spectral leakage corruption to random masking (ablation)SPMAE · K-Radar official splitmAP AP_BEV: 54.3 -> 55.9Improved
leakage corruption alone instead of random masking (ablation)SPMAE · K-Radar official splitmAP AP_BEV: 54.3 -> 53.9No clear effect
use FOCUS power-based masking instead of global random masking (ablation)SPMAE · K-Radar official splitmAP AP_BEV: 54.3 -> 52.1Hurt
stack Doppler as channels (RAE-d) instead of mean Doppler (RAE) (ablation)L-SPMAE · K-Radar official splitmAP AP_BEV: 53.7 -> 52.2Hurt
use RAD (x,y,D) Doppler axis instead of RAE (ablation)L-SPMAE · K-Radar official splitmAP AP_BEV: 53.7 -> 55.9Improved
use radar-only L-SPMAE instead of LiDAR-guided pretraining (ablation)PTv3 · K-Radar official splitmAP AP_3D: 46.8 -> 50.1Improved
train from scratch on L-SPMAE reconstructed spectra (ablation)PTv3 · K-Radar official splitAP_BEV: 48.9 -> 52.0Improved
swap backbone to Sparse Encoder (ablation)L-SPMAE · K-Radar official splitAP_BEV: 47.2 -> 53.1Improved
pretrain with L-SPMAE with 25% labels (ablation)PTv3 · K-Radar official splitmAP AP_BEV: 48.9 (100% scratch) vs 50.4 (25% + L-SPMAE)Improved
add SPMAE/L-SPMAE pretrainingPTv3 · RADIal testmAP: 89.0 -> 90.0Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy