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/

Zendy Plus

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

Zendy Tools

Zendy Open

This paper presents a novel framework for robust 3D object detection frompoint clouds via cross-modal hallucination. Our proposed approach is agnosticto either hallucination direction between LiDAR and 4D radar. We introducemultiple alignments on both spatial and feature levels to achieve simultaneousbackbone refinement and hallucination generation. Specifically, spatialalignment is proposed to deal with the geometry discrepancy for better instancematching between LiDAR and radar. The feature alignment step further bridgesthe intrinsic attribute gap between the sensing modalities and stabilizes thetraining. The trained object detection models can deal with difficult detectioncases better, even though only single-modal data is used as the input duringthe inference stage. Extensive experiments on the View-of-Delft (VoD) datasetshow that our proposed method outperforms the state-of-the-art (SOTA) methodsfor both radar and LiDAR object detection while maintaining competitiveefficiency in runtime. Code is available athttps://github.com/DJNing/See_beyond_seeing.

Robust 3D Object Detection from LiDAR-Radar Point Clouds via Cross-Modal Feature Augmentation