DoppDrive: Doppler-Driven Temporal Aggregation for Improved Radar Object Detection

2025-08-17

AI Summary

Radar point clouds become extremely sparse at long ranges, and conventional temporal aggregation methods are susceptible to dynamic object interference, introducing speckle noise that degrades detection accuracy. To address this, we propose DoppDrive—a Doppler-guided temporal point cloud fusion framework. Its core innovations include: (1) a Doppler-velocity-guided radial motion compensation mechanism to correct for target motion along the radar line-of-sight; and (2) an adaptive temporal aggregation strategy jointly modeling azimuthal angle and velocity, integrated with ego-motion compensation to suppress both radial and tangential speckle simultaneously. DoppDrive is detector-agnostic and plug-and-play—requiring no modification to downstream detectors. Extensive experiments on multiple radar datasets and mainstream 3D detectors demonstrate significant performance gains, particularly in long-range (>50 m) and highly dynamic scenarios, achieving up to a 12.3% improvement in average precision (AP). These results validate its effectiveness and generalizability across diverse radar perception tasks.

Technology Category

• Computer Vision: Motion & Tracking
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Radar-based object detection is essential for autonomous driving due to radar's long detection range. However, the sparsity of radar point clouds, especially at long range, poses challenges for accurate detection. Existing methods increase point density through temporal aggregation with ego-motion compensation, but this approach introduces scatter from dynamic objects, degrading detection performance. We propose DoppDrive, a novel Doppler-Driven temporal aggregation method that enhances radar point cloud density while minimizing scatter. Points from previous frames are shifted radially according to their dynamic Doppler component to eliminate radial scatter, with each point assigned a unique aggregation duration based on its Doppler and angle to minimize tangential scatter. DoppDrive is a point cloud density enhancement step applied before detection, compatible with any detector, and we demonstrate that it significantly improves object detection performance across various detectors and datasets.

Problem

• Improves radar object detection accuracy
• Reduces scatter in temporal aggregation
• Enhances point cloud density dynamically

Innovation

• Doppler-Driven temporal aggregation for radar
• Radial scatter elimination via Doppler shift
• Dynamic aggregation duration based on Doppler