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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

We present SAGNet, a structure-aware generative model for 3D shapes. Given aset of segmented objects of a certain class, the geometry of their parts andthe pairwise relationships between them (the structure) are jointly learned andembedded in a latent space by an autoencoder. The encoder intertwines thegeometry and structure features into a single latent code, while the decoderdisentangles the features and reconstructs the geometry and structure of the 3Dmodel. Our autoencoder consists of two branches, one for the structure and onefor the geometry. The key idea is that during the analysis, the two branchesexchange information between them, thereby learning the dependencies betweenstructure and geometry and encoding two augmented features, which are thenfused into a single latent code. This explicit intertwining of informationenables separately controlling the geometry and the structure of the generatedmodels. We evaluate the performance of our method and conduct an ablationstudy. We explicitly show that encoding of shapes accounts for bothsimilarities in structure and geometry. A variety of quality results generatedby SAGNet are presented. The data and code are athttps://github.com/zhijieW-94/SAGNet.

SAGNet:Structure-aware Generative Network for 3D-Shape Modeling