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

Recent techniques for real-time view synthesis have rapidly advanced infidelity and speed, and modern methods are capable of renderingnear-photorealistic scenes at interactive frame rates. At the same time, atension has arisen between explicit scene representations amenable torasterization and neural fields built on ray marching, with state-of-the-artinstances of the latter surpassing the former in quality while beingprohibitively expensive for real-time applications. In this work, we introduceSMERF, a view synthesis approach that achieves state-of-the-art accuracy amongreal-time methods on large scenes with footprints up to 300 m$^2$ at avolumetric resolution of 3.5 mm$^3$. Our method is built upon two primarycontributions: a hierarchical model partitioning scheme, which increases modelcapacity while constraining compute and memory consumption, and a distillationtraining strategy that simultaneously yields high fidelity and internalconsistency. Our approach enables full six degrees of freedom (6DOF) navigationwithin a web browser and renders in real-time on commodity smartphones andlaptops. Extensive experiments show that our method exceeds the currentstate-of-the-art in real-time novel view synthesis by 0.78 dB on standardbenchmarks and 1.78 dB on large scenes, renders frames three orders ofmagnitude faster than state-of-the-art radiance field models, and achievesreal-time performance across a wide variety of commodity devices, includingsmartphones. We encourage the reader to explore these models in person at ourproject website: https://smerf-3d.github.io.

SMERF: Streamable Memory Efficient Radiance Fields for Real-Time Large-Scene Exploration