English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The 4D Millimeter wave (mmWave) radar is a promising technology for vehiclesensing due to its cost-effectiveness and operability in adverse weatherconditions. However, the adoption of this technology has been hindered bysparsity and noise issues in radar point cloud data. This paper introducesspatial multi-representation fusion (SMURF), a novel approach to 3D objectdetection using a single 4D imaging radar. SMURF leverages multiplerepresentations of radar detection points, including pillarization and densityfeatures of a multi-dimensional Gaussian mixture distribution through kerneldensity estimation (KDE). KDE effectively mitigates measurement inaccuracycaused by limited angular resolution and multi-path propagation of radarsignals. Additionally, KDE helps alleviate point cloud sparsity by capturingdensity features. Experimental evaluations on View-of-Delft (VoD) andTJ4DRadSet datasets demonstrate the effectiveness and generalization ability ofSMURF, outperforming recently proposed 4D imaging radar-basedsingle-representation models. Moreover, while using 4D imaging radar only,SMURF still achieves comparable performance to the state-of-the-art 4D imagingradar and camera fusion-based method, with an increase of 1.22% in the meanaverage precision on bird's-eye view of TJ4DRadSet dataset and 1.32% in the 3Dmean average precision on the entire annotated area of VoD dataset. Ourproposed method demonstrates impressive inference time and addresses thechallenges of real-time detection, with the inference time no more than 0.05seconds for most scans on both datasets. This research highlights the benefitsof 4D mmWave radar and is a strong benchmark for subsequent works regarding 3Dobject detection with 4D imaging radar.

SMURF: Spatial Multi-Representation Fusion for 3D Object Detection with 4D Imaging Radar