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

Pose transfer aims to transfer a given person into a specified posture, hasrecently attracted considerable attention. A typical pose transfer frameworkusually employs representative datasets to train a discriminative model, whichis often violated by out-of-distribution (OOD) instances. Recently, test-timeadaption (TTA) offers a feasible solution for OOD data by using a pre-trainedmodel that learns essential features with self-supervision. However, thosemethods implicitly make an assumption that all test distributions have aunified signal that can be learned directly. In open-world conditions, the posetransfer task raises various independent signals: OOD appearance and skeleton,which need to be extracted and distributed in speciality. To address thispoint, we develop a SEquential Test-time Adaption (SETA). In the test-timephrase, SETA extracts and distributes external appearance texture by augmentingOOD data for self-supervised training. To make non-Euclidean similarity amongdifferent postures explicit, SETA uses the image representations derived from aperson re-identification (Re-ID) model for similarity computation. Byaddressing implicit posture representation in the test-time sequentially, SETAgreatly improves the generalization performance of current pose transfermodels. In our experiment, we first show that pose transfer can be applied toopen-world applications, including Tiktok reenactment and celebrity motionsynthesis.

Open-World Pose Transfer via Sequential Test-Time Adaption