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

Most (3D) multi-object tracking methods rely on appearance-based cues fordata association. By contrast, we investigate how far we can get by onlyencoding geometric relationships between objects in 3D space as cues fordata-driven data association. We encode 3D detections as nodes in a graph,where spatial and temporal pairwise relations among objects are encoded vialocalized polar coordinates on graph edges. This representation makes ourgeometric relations invariant to global transformations and smooth trajectorychanges, especially under non-holonomic motion. This allows our graph neuralnetwork to learn to effectively encode temporal and spatial interactions andfully leverage contextual and motion cues to obtain final scene interpretationby posing data association as edge classification. We establish a newstate-of-the-art on nuScenes dataset and, more importantly, show that ourmethod, PolarMOT, generalizes remarkably well across different locations(Boston, Singapore, Karlsruhe) and datasets (nuScenes and KITTI).

PolarMOT: How Far Can Geometric Relations Take Us in 3D Multi-Object Tracking?