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

Four-dimensional (4D) radar--visual odometry (4DRVO) integrates complementaryinformation from 4D radar and cameras, making it an attractive solution forachieving accurate and robust pose estimation. However, 4DRVO may exhibitsignificant tracking errors owing to three main factors: 1) sparsity of 4Dradar point clouds; 2) inaccurate data association and insufficient featureinteraction between the 4D radar and camera; and 3) disturbances caused bydynamic objects in the environment, affecting odometry estimation. In thispaper, we present 4DRVO-Net, which is a method for 4D radar--visual odometry.This method leverages the feature pyramid, pose warping, and cost volume (PWC)network architecture to progressively estimate and refine poses. Specifically,we propose a multi-scale feature extraction network called Radar-PointNet++that fully considers rich 4D radar point information, enabling fine-grainedlearning for sparse 4D radar point clouds. To effectively integrate the twomodalities, we design an adaptive 4D radar--camera fusion module (A-RCFM) thatautomatically selects image features based on 4D radar point features,facilitating multi-scale cross-modal feature interaction and adaptivemulti-modal feature fusion. In addition, we introduce a velocity-guidedpoint-confidence estimation module to measure local motion patterns, reduce theinfluence of dynamic objects and outliers, and provide continuous updatesduring pose refinement. We demonstrate the excellent performance of our methodand the effectiveness of each module design on both the VoD and in-housedatasets. Our method outperforms all learning-based and geometry-based methodsfor most sequences in the VoD dataset. Furthermore, it has exhibited promisingperformance that closely approaches that of the 64-line LiDAR odometry resultsof A-LOAM without mapping optimization.

4DRVO-Net: Deep 4D Radar-Visual Odometry Using Multi-Modal and Multi-Scale Adaptive Fusion