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

Text-to-3D model adaptations have advanced static 3D model quality, butsequential 3D model generation, particularly for animatable objects with largemotions, is still scarce. Our work proposes AnimatableDreamer, a text-to-4Dgeneration framework capable of generating diverse categories of non-rigidobjects while adhering to the object motions extracted from a monocular video.At its core, AnimatableDreamer is equipped with our novel optimization designdubbed Canonical Score Distillation (CSD), which simplifies the generationdimension from 4D to 3D by denoising over different frames in the time-varyingcamera spaces while conducting the distillation process in a unique canonicalspace shared per video. Concretely, CSD ensures that score gradientsback-propagate to the canonical space through differentiable warping, henceguaranteeing the time-consistent generation and maintaining morphologicalplausibility across different poses. By lifting the 3D generator to 4D withwarping functions, AnimatableDreamer offers a novel perspective on non-rigid 3Dmodel generation and reconstruction. Besides, with inductive knowledge from amulti-view consistent diffusion model, CSD regularizes reconstruction fromnovel views, thus cyclically enhancing the generation process. Extensiveexperiments demonstrate the capability of our method in generatinghigh-flexibility text-guided 3D models from the monocular video, while alsoshowing improved reconstruction performance over typical non-rigidreconstruction methods. Project page https://AnimatableDreamer.github.io.

AnimatableDreamer: Text-Guided Non-rigid 3D Model Generation and Reconstruction with Canonical Score Distillation