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

We address the problem of discovering 3D parts for objects in unseencategories. Being able to learn the geometry prior of parts and transfer thisprior to unseen categories pose fundamental challenges on data-driven shapesegmentation approaches. Formulated as a contextual bandit problem, we proposea learning-based agglomerative clustering framework which learns a groupingpolicy to progressively group small part proposals into bigger ones in abottom-up fashion. At the core of our approach is to restrict the local contextfor extracting part-level features, which encourages the generalizability tounseen categories. On the large-scale fine-grained 3D part dataset, PartNet, wedemonstrate that our method can transfer knowledge of parts learned from 3training categories to 21 unseen testing categories without seeing anyannotated samples. Quantitative comparisons against four shape segmentationbaselines shows that our approach achieve the state-of-the-art performance.

Learning to Group: A Bottom-Up Framework for 3D Part Discovery in Unseen Categories