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 3D scene representations have shown great potential for 3Dreconstruction from 2D images. However, reconstructing real-world captures ofcomplex scenes still remains a challenge. Existing generic 3D reconstructionmethods often struggle to represent fine geometric details and do notadequately model reflective surfaces of large-scale scenes. Techniques thatexplicitly focus on reflective surfaces can model complex and detailedreflections by exploiting better reflection parameterizations. However, weobserve that these methods are often not robust in real unbounded scenarioswhere non-reflective as well as reflective components are present. In thiswork, we propose UniSDF, a general purpose 3D reconstruction method that canreconstruct large complex scenes with reflections. We investigate bothview-based as well as reflection-based color prediction parameterizationtechniques and find that explicitly blending these representations in 3D spaceenables reconstruction of surfaces that are more geometrically accurate,especially for reflective surfaces. We further combine this representation witha multi-resolution grid backbone that is trained in a coarse-to-fine manner,enabling faster reconstructions than prior methods. Extensive experiments onobject-level datasets DTU, Shiny Blender as well as unbounded datasets Mip-NeRF360 and Ref-NeRF real demonstrate that our method is able to robustlyreconstruct complex large-scale scenes with fine details and reflectivesurfaces. Please see our project page athttps://fangjinhuawang.github.io/UniSDF.

UniSDF: Unifying Neural Representations for High-Fidelity 3D Reconstruction of Complex Scenes with Reflections