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

Dynamic Neural Radiance Fields (Dynamic NeRF) enhance NeRF technology tomodel moving scenes. However, they are resource intensive and challenging tocompress. To address this issue, this paper presents WavePlanes, a fast andmore compact explicit model. We propose a multi-scale space and space-timefeature plane representation using N-level 2-D wavelet coefficients. Theinverse discrete wavelet transform reconstructs N feature signals at varyingdetail, which are linearly decoded to approximate the color and density ofvolumes in a 4-D grid. Exploiting the sparsity of wavelet coefficients, wecompress a Hash Map containing only non-zero coefficients and their locationson each plane. This results in a compressed model size of ~12 MB. Compared withstate-of-the-art plane-based models, WavePlanes is up to 15x smaller, lesscomputationally demanding and achieves comparable results in as little as onehour of training - without requiring custom CUDA code or high performancecomputing resources. Additionally, we propose new feature fusion schemes thatwork as well as previously proposed schemes while providing greaterinterpretability. Our code is available at:https://github.com/azzarelli/waveplanes/

WavePlanes: A compact Wavelet representation for Dynamic Neural Radiance Fields