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

Emerging neural radiance fields (NeRF) are a promising scene representationfor computer graphics, enabling high-quality 3D reconstruction and novel viewsynthesis from image observations. However, editing a scene represented by aNeRF is challenging, as the underlying connectionist representations such asMLPs or voxel grids are not object-centric or compositional. In particular, ithas been difficult to selectively edit specific regions or objects. In thiswork, we tackle the problem of semantic scene decomposition of NeRFs to enablequery-based local editing of the represented 3D scenes. We propose to distillthe knowledge of off-the-shelf, self-supervised 2D image feature extractorssuch as CLIP-LSeg or DINO into a 3D feature field optimized in parallel to theradiance field. Given a user-specified query of various modalities such astext, an image patch, or a point-and-click selection, 3D feature fieldssemantically decompose 3D space without the need for re-training and enable usto semantically select and edit regions in the radiance field. Our experimentsvalidate that the distilled feature fields (DFFs) can transfer recent progressin 2D vision and language foundation models to 3D scene representations,enabling convincing 3D segmentation and selective editing of emerging neuralgraphics representations.

Decomposing NeRF for Editing via Feature Field Distillation