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 introduce CAPRI-Net, a neural network for learning compact andinterpretable implicit representations of 3D computer-aided design (CAD)models, in the form of adaptive primitive assemblies. Our network takes aninput 3D shape that can be provided as a point cloud or voxel grids, andreconstructs it by a compact assembly of quadric surface primitives viaconstructive solid geometry (CSG) operations. The network is self-supervisedwith a reconstruction loss, leading to faithful 3D reconstructions with sharpedges and plausible CSG trees, without any ground-truth shape assemblies. Whilethe parametric nature of CAD models does make them more predictable locally, atthe shape level, there is a great deal of structural and topologicalvariations, which present a significant generalizability challenge tostate-of-the-art neural models for 3D shapes. Our network addresses thischallenge by adaptive training with respect to each test shape, with which wefine-tune the network that was pre-trained on a model collection. We evaluateour learning framework on both ShapeNet and ABC, the largest and most diverseCAD dataset to date, in terms of reconstruction quality, shape edges,compactness, and interpretability, to demonstrate superiority over currentalternatives suitable for neural CAD reconstruction.

CAPRI-Net: Learning Compact CAD Shapes with Adaptive Primitive Assembly