HyperDet: 3D Object Detection with Hyper 4D Radar Point Clouds

TL;DR

HyperDet is a novel radar-only 3D detection framework that enhances sparse radar point clouds by aggregating multi-view data, validating geometry, and densifying object structures, significantly narrowing the gap with LiDAR-based systems.

Contribution

It introduces a task-aware hyper 4D radar point cloud construction and a fusion method that improves radar detection accuracy without changing existing detectors.

Findings

HyperDet improves detection accuracy over raw radar inputs.

It narrows the gap between radar and LiDAR detection performance.

The method enhances radar data quality for standard LiDAR-oriented detectors.

Abstract

4D mmWave radar provides weather-robust, velocity-aware measurements and is more cost-effective than LiDAR. However, radar-only 3D detection still trails LiDAR-based systems because radar point clouds are sparse, irregular, and often corrupted by multipath noise, yielding weak and unstable geometry. We present HyperDet, a detector-agnostic radar-only 3D detection framework that constructs a task-aware hyper 4D radar point cloud for standard LiDAR-oriented detectors. HyperDet aggregates returns from multiple surround-view 4D radars over consecutive frames to improve coverage and density, then applies geometry-aware cross-sensor consensus validation with a lightweight self-consistency check outside overlap regions to suppress inconsistent returns. It further integrates a foreground-focused diffusion module with training-time mixed radar-LiDAR supervision to densify object structures while lifting radar attributes (e.g., Doppler, RCS); the model is distilled into a consistency model for single-step inference. On MAN TruckScenes, HyperDet consistently improves over raw radar inputs with VoxelNeXt and CenterPoint, partially narrowing the radar-LiDAR gap. These results show that input-level refinement enables radar to better leverage LiDAR-oriented detectors without architectural modifications.