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 task of 3D single object tracking (SOT) with LiDAR point clouds iscrucial for various applications, such as autonomous driving and robotics.However, existing approaches have primarily relied on appearance matching ormotion modeling within only two successive frames, thereby overlooking thelong-range continuous motion property of objects in 3D space. To address thisissue, this paper presents a novel approach that views each tracklet as acontinuous stream: at each timestamp, only the current frame is fed into thenetwork to interact with multi-frame historical features stored in a memorybank, enabling efficient exploitation of sequential information. To achieveeffective cross-frame message passing, a hybrid attention mechanism is designedto account for both long-range relation modeling and local geometric featureextraction. Furthermore, to enhance the utilization of multi-frame features forrobust tracking, a contrastive sequence enhancement strategy is designed, whichuses ground truth tracklets to augment training sequences and promotediscrimination against false positives in a contrastive manner. Extensiveexperiments demonstrate that the proposed method outperforms thestate-of-the-art method by significant margins (approximately 8%, 6%, and 12%improvements in the success performance on KITTI, nuScenes, and Waymo,respectively).

Modeling Continuous Motion for 3D Point Cloud Object Tracking