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

Cross-scene generalizable NeRF models, which can directly synthesize novelviews of unseen scenes, have become a new spotlight of the NeRF field. Severalexisting attempts rely on increasingly end-to-end "neuralized" architectures,i.e., replacing scene representation and/or rendering modules with performantneural networks such as transformers, and turning novel view synthesis into afeed-forward inference pipeline. While those feedforward "neuralized"architectures still do not fit diverse scenes well out of the box, we proposeto bridge them with the powerful Mixture-of-Experts (MoE) idea from largelanguage models (LLMs), which has demonstrated superior generalization abilityby balancing between larger overall model capacity and flexible per-instancespecialization. Starting from a recent generalizable NeRF architecture calledGNT, we first demonstrate that MoE can be neatly plugged in to enhance themodel. We further customize a shared permanent expert and a geometry-awareconsistency loss to enforce cross-scene consistency and spatial smoothnessrespectively, which are essential for generalizable view synthesis. Ourproposed model, dubbed GNT with Mixture-of-View-Experts (GNT-MOVE), hasexperimentally shown state-of-the-art results when transferring to unseenscenes, indicating remarkably better cross-scene generalization in bothzero-shot and few-shot settings. Our codes are available athttps://github.com/VITA-Group/GNT-MOVE.

Enhancing NeRF akin to Enhancing LLMs: Generalizable NeRF Transformer with Mixture-of-View-Experts