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

In this paper, we present a deep learning model that exploits the power ofself-supervision to perform 3D point cloud completion, estimating the missingpart and a context region around it. Local and global information are encodedin a combined embedding. A denoising pretext task provides the network with theneeded local cues, decoupled from the high-level semantics and naturally sharedover multiple classes. On the other hand, contrastive learning maximizes theagreement between variants of the same shape with different missing portions,thus producing a representation which captures the global appearance of theshape. The combined embedding inherits category-agnostic properties from thechosen pretext tasks. Differently from existing approaches, this allows tobetter generalize the completion properties to new categories unseen attraining time. Moreover, while decoding the obtained joint representation, webetter blend the reconstructed missing part with the partial shape by payingattention to its known surrounding region and reconstructing this frame asauxiliary objective. Our extensive experiments and detailed ablation on theShapeNet dataset show the effectiveness of each part of the method with newstate of the art results. Our quantitative and qualitative analysis confirmshow our approach is able to work on novel categories without relying neither onclassification and shape symmetry priors, nor on adversarial trainingprocedures.

Denoise and Contrast for Category Agnostic Shape Completion