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

3D object recognition has seen significant advances in recent years, showingimpressive performance on real-world 3D scan benchmarks, but lacking in objectpart reasoning, which is fundamental to higher-level scene understanding suchas inter-object similarities or object functionality. Thus, we propose toleverage large-scale synthetic datasets of 3D shapes annotated with partinformation to learn Neural Part Priors (NPPs), optimizable spacescharacterizing geometric part priors. Crucially, we can optimize over thelearned part priors in order to fit to real-world scanned 3D scenes at testtime, enabling robust part decomposition of the real objects in these scenesthat also estimates the complete geometry of the object while fittingaccurately to the observed real geometry. Moreover, this enables globaloptimization over geometrically similar detected objects in a scene, whichoften share strong geometric commonalities, enabling scene-consistent partdecompositions. Experiments on the ScanNet dataset demonstrate that NPPssignificantly outperforms state of the art in part decomposition and objectcompletion in real-world scenes.

Neural Part Priors: Learning to Optimize Part-Based Object Completion in RGB-D Scans