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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

We, as human beings, can understand and picture a familiar scene fromarbitrary viewpoints given a single image, whereas this is still a grandchallenge for computers. We hereby present a novel solution to mimic such humanperception capability based on a new paradigm of amodal 3D scene understandingwith neural rendering for a closed scene. Specifically, we first learn theprior knowledge of the objects in a closed scene via an offline stage, whichfacilitates an online stage to understand the room with unseen furniturearrangement. During the online stage, given a panoramic image of the scene indifferent layouts, we utilize a holistic neural-rendering-based optimizationframework to efficiently estimate the correct 3D scene layout and deliverrealistic free-viewpoint rendering. In order to handle the domain gap betweenthe offline and online stage, our method exploits compositional neuralrendering techniques for data augmentation in the offline training. Theexperiments on both synthetic and real datasets demonstrate that our two-stagedesign achieves robust 3D scene understanding and outperforms competing methodsby a large margin, and we also show that our realistic free-viewpoint renderingenables various applications, including scene touring and editing. Code anddata are available on the project webpage:https://zju3dv.github.io/nr_in_a_room/.

Neural Rendering in a Room: Amodal 3D Understanding and Free-Viewpoint Rendering for the Closed Scene Composed of Pre-Captured Objects