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 Tools annual

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 monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Neural implicit representations are known to be more compact for depicting 3Dshapes than traditional discrete representations. However, the neuralrepresentations tend to round sharp corners or edges and struggle to representsurfaces with open boundaries. Moreover, they are slow to train. We present aunified neural implicit representation, called Patch-Grid, that fits to complexshapes efficiently, preserves sharp features, and effectively models surfaceswith open boundaries and thin geometric features. Our superior efficiency comesfrom embedding each surface patch into a local latent volume and decoding itusing a shared MLP decoder, which is pretrained on various local surfacegeometries. With this pretrained decoder fixed, fitting novel shapes and localshape updates can be done efficiently. The faithful preservation of sharpfeatures is enabled by adopting a novel merge grid to perform localconstructive solid geometry (CSG) combinations of surface patches in the cellsof an adaptive Octree, yielding better robustness than using a global CSGconstruction as proposed in the literature. Experiments show that ourPatch-Grid method faithfully captures shapes with complex sharp features, openboundaries and thin structures, and outperforms existing learning-based methodsin both efficiency and quality for surface fitting and local shape updates.

Patch-Grid: An Efficient and Feature-Preserving Neural Implicit Surface Representation