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 focus on the problem of novel-view human action synthesis. Given an actionvideo, the goal is to generate the same action from an unseen viewpoint.Naturally, novel view video synthesis is more challenging than image synthesis.It requires the synthesis of a sequence of realistic frames with temporalcoherency. Besides, transferring the different actions to a novel target viewrequires awareness of action category and viewpoint change simultaneously. Toaddress these challenges, we propose a novel framework named Pose-guided ActionSeparable Generative Adversarial Net (PAS-GAN), which utilizes pose toalleviate the difficulty of this task. First, we propose a recurrentpose-transformation module which transforms actions from the source view to thetarget view and generates novel view pose sequence in 2D coordinate space.Second, a well-transformed pose sequence enables us to separatethe action andbackground in the target view. We employ a novel local-global spatialtransformation module to effectively generate sequential video features in thetarget view using these action and background features. Finally, the generatedvideo features are used to synthesize human action with the help of a 3Ddecoder. Moreover, to focus on dynamic action in the video, we propose a novelmulti-scale action-separable loss which further improves the video quality. Weconduct extensive experiments on two large-scale multi-view human actiondatasets, NTU-RGBD and PKU-MMD, demonstrating the effectiveness of PAS-GANwhich outperforms existing approaches.

Pose-guided Generative Adversarial Net for Novel View Action Synthesis