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

Implicit Neural Representations (INRs) have recently exhibited immensepotential in the field of scientific visualization for both data generation andvisualization tasks. However, these representations often consist of largemulti-layer perceptrons (MLPs), necessitating millions of operations for asingle forward pass, consequently hindering interactive visual exploration.While reducing the size of the MLPs and employing efficient parametric encodingschemes can alleviate this issue, it compromises generalizability for unseenparameters, rendering it unsuitable for tasks such as temporalsuper-resolution. In this paper, we introduce HyperINR, a novel hypernetworkarchitecture capable of directly predicting the weights for a compact INR. Byharnessing an ensemble of multiresolution hash encoding units in unison, theresulting INR attains state-of-the-art inference performance (up to 100x higherinference bandwidth) and can support interactive photo-realistic volumevisualization. Additionally, by incorporating knowledge distillation,exceptional data and visualization generation quality is achieved, making ourmethod valuable for real-time parameter exploration. We validate theeffectiveness of the HyperINR architecture through a comprehensive ablationstudy. We showcase the versatility of HyperINR across three distinct scientificdomains: novel view synthesis, temporal super-resolution of volume data, andvolume rendering with dynamic global shadows. By simultaneously achievingefficiency and generalizability, HyperINR paves the way for applying INR in awider array of scientific visualization applications.

HyperINR: A Fast and Predictive Hypernetwork for Implicit Neural Representations via Knowledge Distillation