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

Synthesizing novel view images from a few views is a challenging butpractical problem. Existing methods often struggle with producing high-qualityresults or necessitate per-object optimization in such few-view settings due tothe insufficient information provided. In this work, we explore leveraging thestrong 2D priors in pre-trained diffusion models for synthesizing novel viewimages. 2D diffusion models, nevertheless, lack 3D awareness, leading todistorted image synthesis and compromising the identity. To address theseproblems, we propose DreamSparse, a framework that enables the frozenpre-trained diffusion model to generate geometry and identity-consistent novelview image. Specifically, DreamSparse incorporates a geometry module designedto capture 3D features from sparse views as a 3D prior. Subsequently, a spatialguidance model is introduced to convert these 3D feature maps into spatialinformation for the generative process. This information is then used to guidethe pre-trained diffusion model, enabling it to generate geometricallyconsistent images without tuning it. Leveraging the strong image priors in thepre-trained diffusion models, DreamSparse is capable of synthesizinghigh-quality novel views for both object and scene-level images andgeneralising to open-set images. Experimental results demonstrate that ourframework can effectively synthesize novel view images from sparse views andoutperforms baselines in both trained and open-set category images. Moreresults can be found on our project page:https://sites.google.com/view/dreamsparse-webpage.

DreamSparse: Escaping from Plato's Cave with 2D Frozen Diffusion Model Given Sparse Views