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

The task of unsupervised motion retargeting in videos has seen substantialadvancements through the use of deep neural networks. While early worksconcentrated on specific object priors such as a human face or body, recentwork considered the unsupervised case. When the source and target videos,however, are of different shapes, current methods fail. To alleviate thisproblem, we introduce JOKR - a JOint Keypoint Representation that captures themotion common to both the source and target videos, without requiring anyobject prior or data collection. By employing a domain confusion term, weenforce the unsupervised keypoint representations of both videos to beindistinguishable. This encourages disentanglement between the parts of themotion that are common to the two domains, and their distinctive appearance andmotion, enabling the generation of videos that capture the motion of the onewhile depicting the style of the other. To enable cases where the objects areof different proportions or orientations, we apply a learned affinetransformation between the JOKRs. This augments the representation to be affineinvariant, and in practice broadens the variety of possible retargeting pairs.This geometry-driven representation enables further intuitive control, such astemporal coherence and manual editing. Through comprehensive experimentation,we demonstrate the applicability of our method to different challengingcross-domain video pairs. We evaluate our method both qualitatively andquantitatively, and demonstrate that our method handles various cross-domainscenarios, such as different animals, different flowers, and humans. We alsodemonstrate superior temporal coherency and visual quality compared tostate-of-the-art alternatives, through statistical metrics and a user study.Source code and videos can be found at https://rmokady.github.io/JOKR/ .

JOKR: Joint Keypoint Representation for Unsupervised Cross-Domain Motion Retargeting