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/

Zendy Plus

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

Zendy Tools

Zendy Open

We present Neural 3D Strokes, a novel technique to generate stylized imagesof a 3D scene at arbitrary novel views from multi-view 2D images. Differentfrom existing methods which apply stylization to trained neural radiance fieldsat the voxel level, our approach draws inspiration from image-to-paintingmethods, simulating the progressive painting process of human artwork withvector strokes. We develop a palette of stylized 3D strokes from basicprimitives and splines, and consider the 3D scene stylization task as amulti-view reconstruction process based on these 3D stroke primitives. Insteadof directly searching for the parameters of these 3D strokes, which would betoo costly, we introduce a differentiable renderer that allows optimizingstroke parameters using gradient descent, and propose a training scheme toalleviate the vanishing gradient issue. The extensive evaluation demonstratesthat our approach effectively synthesizes 3D scenes with significant geometricand aesthetic stylization while maintaining a consistent appearance acrossdifferent views. Our method can be further integrated with style loss andimage-text contrastive models to extend its applications, including colortransfer and text-driven 3D scene drawing. Results and code are available athttp://buaavrcg.github.io/Neural3DStrokes.

Neural 3D Strokes: Creating Stylized 3D Scenes with Vectorized 3D Strokes