AI Paper Atlas

Research map / Detection and segmentation / 3D detection and driving

HVA-Fusion:Hierarchical Velocity-Aware 4D Radar-LiDAR Fusion for Robust 3D Object Detection

Jingxian Wu, Lu Wang, Lisheng Xu, Jun Cheng

ECCV 2026 · Detection and segmentation › 3D detection and driving · approach: Camera-radar fusion detection · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add MVEE (true velocity estimation) to L4DR baseline (ablation)L4DR · VoD Entire AreamAP: 70.99 -> 73.36Improved
add MVEE (ablation)L4DR · VoD Driving AreamAP: 83.95 -> 85.35Improved
add CPA on top of MVEE (ablation)HVA-Fusion · VoD Entire AreamAP: 73.36 -> 73.71Improved
add CPA on top of MVEE (ablation)HVA-Fusion · VoD Driving AreamAP: 85.35 -> 88.14Improved
add RGG on top of MVEE+CPA (ablation)HVA-Fusion · VoD Entire AreamAP: 73.71 -> 74.11Improved
add RGG to full model (ablation)HVA-Fusion · VoD Entire AreamAP: 74.77 -> 75.51Improved
add CPA to full model (ablation)HVA-Fusion · VoD Entire AreamAP: 74.20 -> 75.51Improved
add MS-BDDG to form full model (ablation)HVA-Fusion · VoD Entire AreamAP: 74.11 -> 75.51Improved
disable CPA at inference under extreme fog (ablation)HVA-Fusion · VoD-Fog level 4Entire Area mAP: -1.76%Hurt
reduce CPA interaction strength under extreme fog (ablation)HVA-Fusion · VoD-Fog level 4Entire Area mAP: +0.31%Improved
use HVA-Fusion instead of L4DRfusion · VoD-Fog level 4mAP: 28.0 -> 25.8Hurt
use HVA-Fusion instead of L4DRfusion · K-RadarAP_3D total: 53.5 -> 65.7Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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