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

Neural Radiance Fields (NeRF) has achieved unprecedented view synthesisquality using coordinate-based neural scene representations. However, NeRF'sview dependency can only handle simple reflections like highlights but cannotdeal with complex reflections such as those from glass and mirrors. In thesescenarios, NeRF models the virtual image as real geometries which leads toinaccurate depth estimation, and produces blurry renderings when the multi-viewconsistency is violated as the reflected objects may only be seen under some ofthe viewpoints. To overcome these issues, we introduce NeRFReN, which is builtupon NeRF to model scenes with reflections. Specifically, we propose to split ascene into transmitted and reflected components, and model the two componentswith separate neural radiance fields. Considering that this decomposition ishighly under-constrained, we exploit geometric priors and applycarefully-designed training strategies to achieve reasonable decompositionresults. Experiments on various self-captured scenes show that our methodachieves high-quality novel view synthesis and physically sound depthestimation results while enabling scene editing applications.

NeRFReN: Neural Radiance Fields with Reflections