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

This paper proposes a Graph Neural Network(GNN)-based method for exploitingsemantics and local geometry to guide the identification of reliable pointcloudregistration candidates. Semantic and morphological features of the environmentserve as key reference points for registration, enabling accurate lidar-basedpose estimation. Our novel lightweight static graph structure informs ourattention-based node aggregation network by identifying semantic-instancerelationships, acting as an inductive bias to significantly reduce thecomputational burden of pointcloud registration. By connecting candidate nodesand exploiting cross-graph attention, we identify confidence scores for allpotential registration correspondences and estimate the displacement betweenpointcloud scans. Our pipeline enables introspective analysis of the model'sperformance by correlating it with the individual contributions of localstructures in the environment, providing valuable insights into the system'sbehaviour. We test our method on the KITTI odometry dataset, achievingcompetitive accuracy compared to benchmark methods and a higher tracksmoothness while relying on significantly fewer network parameters.

SEM-GAT: Explainable Semantic Pose Estimation using Learned Graph Attention