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

A robust and accurate 3D detection system is an integral part of autonomousvehicles. Traditionally, a majority of 3D object detection algorithms focus onprocessing 3D point clouds using voxel grids or bird's eye view (BEV). Recentworks, however, demonstrate the utilization of the graph neural network (GNN)as a promising approach to 3D object detection. In this work, we propose anattention based feature aggregation technique in GNN for detecting objects inLiDAR scan. We first employ a distance-aware down-sampling scheme that not onlyenhances the algorithmic performance but also retains maximum geometricfeatures of objects even if they lie far from the sensor. In each layer of theGNN, apart from the linear transformation which maps the per node inputfeatures to the corresponding higher level features, a per node maskedattention by specifying different weights to different nodes in its first ringneighborhood is also performed. The masked attention implicitly accounts forthe underlying neighborhood graph structure of every node and also eliminatesthe need of costly matrix operations thereby improving the detection accuracywithout compromising the performance. The experiments on KITTI dataset showthat our method yields comparable results for 3D object detection.

Dynamic Edge Weights in Graph Neural Networks for 3D Object Detection