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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Neural Radiance Fields (NeRF) can generate highly realistic novel views.However, editing 3D scenes represented by NeRF across 360-degree views,particularly removing objects while preserving geometric and photometricconsistency, remains a challenging problem due to NeRF's implicit scenerepresentation. In this paper, we propose InpaintNeRF360, a unified frameworkthat utilizes natural language instructions as guidance for inpaintingNeRF-based 3D scenes.Our approach employs a promptable segmentation model bygenerating multi-modal prompts from the encoded text for multiviewsegmentation. We apply depth-space warping to enforce viewing consistency inthe segmentations, and further refine the inpainted NeRF model using perceptualpriors to ensure visual plausibility. InpaintNeRF360 is capable ofsimultaneously removing multiple objects or modifying object appearance basedon text instructions while synthesizing 3D viewing-consistent andphoto-realistic inpainting. Through extensive experiments on both unbounded andfrontal-facing scenes trained through NeRF, we demonstrate the effectiveness ofour approach and showcase its potential to enhance the editability of implicitradiance fields.

InpaintNeRF360: Text-Guided 3D Inpainting on Unbounded Neural Radiance Fields