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/

Zendy Plus

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

Zendy Tools

Zendy Open

Neural shape representation generally refers to representing 3D geometryusing neural networks, e.g., computing a signed distance or occupancy value ata specific spatial position. In this paper we present a neural-networkarchitecture suitable for accurate encoding of 3D shapes in a single forwardpass. Our architecture is based on a multi-scale hybrid system incorporatinggraph-based and voxel-based components, as well as a continuouslydifferentiable decoder. The hybrid system includes a novel way of voxelizingpoint-based features in neural networks, which we show can be used incombination with oriented point-clouds to obtain smoother and more detailedreconstructions. Furthermore, our network is trained to solve the eikonalequation and only requires knowledge of the zero-level set for training andinference. This means that in contrast to most previous shape encoderarchitectures, our network is able to output valid signed distance fieldswithout explicit prior knowledge of non-zero distance values or shapeoccupancy. It also requires only a single forward-pass, instead of thelatent-code optimization used in auto-decoder methods. We further propose amodification to the loss function in case that surface normals are not welldefined, e.g., in the context of non-watertight surfaces and non-manifoldgeometry, resulting in an unsigned distance field. Overall, our system can helpto reduce the computational overhead of training and evaluating neural distancefields, as well as enabling the application to difficult geometry.

HYVE: Hybrid Vertex Encoder for Neural Distance Fields