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

We introduce Ponymation, a new method for learning a generative model ofarticulated 3D animal motions from raw, unlabeled online videos. Unlikeexisting approaches for motion synthesis, our model does not require any poseannotations or parametric shape models for training, and is learned purely froma collection of raw video clips obtained from the Internet. We build upon arecent work, MagicPony, which learns articulated 3D animal shapes purely fromsingle image collections, and extend it on two fronts. First, instead oftraining on static images, we augment the framework with a video trainingpipeline that incorporates temporal regularizations, achieving more accurateand temporally consistent reconstructions. Second, we learn a generative modelof the underlying articulated 3D motion sequences via a spatio-temporaltransformer VAE, simply using 2D reconstruction losses without relying on anyexplicit pose annotations. At inference time, given a single 2D image of a newanimal instance, our model reconstructs an articulated, textured 3D mesh, andgenerates plausible 3D animations by sampling from the learned motion latentspace.

Ponymation: Learning 3D Animal Motions from Unlabeled Online Videos