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

Learning deformable 3D objects from 2D images is often an ill-posed problem.Existing methods rely on explicit supervision to establish multi-viewcorrespondences, such as template shape models and keypoint annotations, whichrestricts their applicability on objects "in the wild". A more natural way ofestablishing correspondences is by watching videos of objects moving around. Inthis paper, we present DOVE, a method that learns textured 3D models ofdeformable object categories from monocular videos available online, withoutkeypoint, viewpoint or template shape supervision. By resolvingsymmetry-induced pose ambiguities and leveraging temporal correspondences invideos, the model automatically learns to factor out 3D shape, articulated poseand texture from each individual RGB frame, and is ready for single-imageinference at test time. In the experiments, we show that existing methods failto learn sensible 3D shapes without additional keypoint or templatesupervision, whereas our method produces temporally consistent 3D models, whichcan be animated and rendered from arbitrary viewpoints.

DOVE: Learning Deformable 3D Objects by Watching Videos