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

Fourier PlenOctrees have shown to be an efficient representation forreal-time rendering of dynamic Neural Radiance Fields (NeRF). Despite its manyadvantages, this method suffers from artifacts introduced by the involvedcompression when combining it with recent state-of-the-art techniques fortraining the static per-frame NeRF models. In this paper, we perform anin-depth analysis of these artifacts and leverage the resulting insights topropose an improved representation. In particular, we present a novel densityencoding that adapts the Fourier-based compression to the characteristics ofthe transfer function used by the underlying volume rendering procedure andleads to a substantial reduction of artifacts in the dynamic model.Furthermore, we show an augmentation of the training data that relaxes theperiodicity assumption of the compression. We demonstrate the effectiveness ofour enhanced Fourier PlenOctrees in the scope of quantitative and qualitativeevaluations on synthetic and real-world scenes.

FPO++: Efficient Encoding and Rendering of Dynamic Neural Radiance Fields by Analyzing and Enhancing Fourier PlenOctrees