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

Existing LiDAR-based 3D object detectors usually focus on the single-framedetection, while ignoring the spatiotemporal information in consecutive pointcloud frames. In this paper, we propose an end-to-end online 3D video objectdetector that operates on point cloud sequences. The proposed model comprises aspatial feature encoding component and a spatiotemporal feature aggregationcomponent. In the former component, a novel Pillar Message Passing Network(PMPNet) is proposed to encode each discrete point cloud frame. It adaptivelycollects information for a pillar node from its neighbors by iterative messagepassing, which effectively enlarges the receptive field of the pillar feature.In the latter component, we propose an Attentive Spatiotemporal Transformer GRU(AST-GRU) to aggregate the spatiotemporal information, which enhances theconventional ConvGRU with an attentive memory gating mechanism. AST-GRUcontains a Spatial Transformer Attention (STA) module and a TemporalTransformer Attention (TTA) module, which can emphasize the foreground objectsand align the dynamic objects, respectively. Experimental results demonstratethat the proposed 3D video object detector achieves state-of-the-artperformance on the large-scale nuScenes benchmark.

LiDAR-based Online 3D Video Object Detection with Graph-based Message Passing and Spatiotemporal Transformer Attention