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

High fidelity representation of shapes with arbitrary topology is animportant problem for a variety of vision and graphics applications. Owing totheir limited resolution, classical discrete shape representations using pointclouds, voxels and meshes produce low quality results when used in theseapplications. Several implicit 3D shape representation approaches using deepneural networks have been proposed leading to significant improvements in bothquality of representations as well as the impact on downstream applications.However, these methods can only be used to represent topologically closedshapes which greatly limits the class of shapes that they can represent. As aconsequence, they also often require clean, watertight meshes for training. Inthis work, we propose DUDE - a Deep Unsigned Distance Embedding method whichalleviates both of these shortcomings. DUDE is a disentangled shaperepresentation that utilizes an unsigned distance field (uDF) to representproximity to a surface, and a normal vector field (nVF) to represent surfaceorientation. We show that a combination of these two (uDF+nVF) can be used tolearn high fidelity representations for arbitrary open/closed shapes. Asopposed to prior work such as DeepSDF, our shape representations can bedirectly learnt from noisy triangle soups, and do not need watertight meshes.Additionally, we propose novel algorithms for extracting and renderingiso-surfaces from the learnt representations. We validate DUDE on benchmark 3Ddatasets and demonstrate that it produces significant improvements over thestate of the art.

DUDE: Deep Unsigned Distance Embeddings for Hi-Fidelity Representation of Complex 3D Surfaces