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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

An unsupervised human action modeling framework can provide usefulpose-sequence representation, which can be utilized in a variety of poseanalysis applications. In this work we propose a novel temporal pose-sequencemodeling framework, which can embed the dynamics of 3D human-skeleton joints toa continuous latent space in an efficient manner. In contrast to end-to-endframework explored by previous works, we disentangle the task of individualpose representation learning from the task of learning actions as a trajectoryin pose embedding space. In order to realize a continuous pose embeddingmanifold with improved reconstructions, we propose an unsupervised, manifoldlearning procedure named Encoder GAN, (or EnGAN). Further, we use the poseembeddings generated by EnGAN to model human actions using a bidirectional RNNauto-encoder architecture, PoseRNN. We introduce first-order gradient loss toexplicitly enforce temporal regularity in the predicted motion sequence. Ahierarchical feature fusion technique is also investigated for simultaneousmodeling of local skeleton joints along with global pose variations. Wedemonstrate state-of-the-art transfer-ability of the learned representationagainst other supervisedly and unsupervisedly learned motion embeddings for thetask of fine-grained action recognition on SBU interaction dataset. Further, weshow the qualitative strengths of the proposed framework by visualizingskeleton pose reconstructions and interpolations in pose-embedding space, andlow dimensional principal component projections of the reconstructed posetrajectories.

Unsupervised Feature Learning of Human Actions as Trajectories in Pose Embedding Manifold