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

In this paper, we aim to learn a semantic radiance field from multiple scenesthat is accurate, efficient and generalizable. While most existing NeRFs targetat the tasks of neural scene rendering, image synthesis and multi-viewreconstruction, there are a few attempts such as Semantic-NeRF that explore tolearn high-level semantic understanding with the NeRF structure. However,Semantic-NeRF simultaneously learns color and semantic label from a single raywith multiple heads, where the single ray fails to provide rich semanticinformation. As a result, Semantic NeRF relies on positional encoding and needsto train one specific model for each scene. To address this, we proposeSemantic Ray (S-Ray) to fully exploit semantic information along the raydirection from its multi-view reprojections. As directly performing denseattention over multi-view reprojected rays would suffer from heavycomputational cost, we design a Cross-Reprojection Attention module withconsecutive intra-view radial and cross-view sparse attentions, whichdecomposes contextual information along reprojected rays and cross multipleviews and then collects dense connections by stacking the modules. Experimentsshow that our S-Ray is able to learn from multiple scenes, and it presentsstrong generalization ability to adapt to unseen scenes.

Semantic Ray: Learning a Generalizable Semantic Field with Cross-Reprojection Attention