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

Boundary Representations (B-Reps) are the industry standard in 3D ComputerAided Design/Manufacturing (CAD/CAM) and industrial design due to theirfidelity in representing stylistic details. However, they have been ignored inthe 3D style research. Existing 3D style metrics typically operate on meshes orpointclouds, and fail to account for end-user subjectivity by adopting fixeddefinitions of style, either through crowd-sourcing for style labels orhand-crafted features. We propose UVStyle-Net, a style similarity measure forB-Reps that leverages the style signals in the second order statistics of theactivations in a pre-trained (unsupervised) 3D encoder, and learns theirrelative importance to a subjective end-user through few-shot learning. Ourapproach differs from all existing data-driven 3D style methods since it may beused in completely unsupervised settings, which is desirable given the lack ofpublicly available labelled B-Rep datasets. More importantly, the few-shotlearning accounts for the inherent subjectivity associated with style. We showquantitatively that our proposed method with B-Reps is able to capture strongerstyle signals than alternative methods on meshes and pointclouds despite itssignificantly greater computational efficiency. We also show it is able togenerate meaningful style gradients with respect to the input shape, and thatfew-shot learning with as few as two positive examples selected by an end-useris sufficient to significantly improve the style measure. Finally, wedemonstrate its efficacy on a large unlabeled public dataset of CAD models.Source code and data will be released in the future.

UVStyle-Net: Unsupervised Few-shot Learning of 3D Style Similarity Measure for B-Reps