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

Reconstructing 3D objects from a single image guided by pretrained diffusionmodels has demonstrated promising outcomes. However, due to utilizing thecase-agnostic rigid strategy, their generalization ability to arbitrary casesand the 3D consistency of reconstruction are still poor. In this work, wepropose Consistent123, a case-aware two-stage method for highly consistent 3Dasset reconstruction from one image with both 2D and 3D diffusion priors. Inthe first stage, Consistent123 utilizes only 3D structural priors forsufficient geometry exploitation, with a CLIP-based case-aware adaptivedetection mechanism embedded within this process. In the second stage, 2Dtexture priors are introduced and progressively take on a dominant guidingrole, delicately sculpting the details of the 3D model. Consistent123 alignsmore closely with the evolving trends in guidance requirements, adaptivelyproviding adequate 3D geometric initialization and suitable 2D texturerefinement for different objects. Consistent123 can obtain highly 3D-consistentreconstruction and exhibits strong generalization ability across variousobjects. Qualitative and quantitative experiments show that our methodsignificantly outperforms state-of-the-art image-to-3D methods. Seehttps://Consistent123.github.io for a more comprehensive exploration of ourgenerated 3D assets.

Consistent123: One Image to Highly Consistent 3D Asset Using Case-Aware Diffusion Priors