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

This paper addresses the problem of 2D pose representation duringunsupervised 2D to 3D pose lifting to improve the accuracy, stability andgeneralisability of 3D human pose estimation (HPE) models. All unsupervised2D-3D HPE approaches provide the entire 2D kinematic skeleton to a model duringtraining. We argue that this is sub-optimal and disruptive as long-rangecorrelations are induced between independent 2D key points and predicted 3Dordinates during training. To this end, we conduct the following study. With amaximum architecture capacity of 6 residual blocks, we evaluate the performanceof 5 models which each represent a 2D pose differently during the adversarialunsupervised 2D-3D HPE process. Additionally, we show the correlations between2D key points which are learned during the training process, highlighting theunintuitive correlations induced when an entire 2D pose is provided to alifting model. Our results show that the most optimal representation of a 2Dpose is that of two independent segments, the torso and legs, with no sharedfeatures between each lifting network. This approach decreased the averageerror by 20\% on the Human3.6M dataset when compared to a model with a nearidentical parameter count trained on the entire 2D kinematic skeleton.Furthermore, due to the complex nature of adversarial learning, we show howthis representation can also improve convergence during training allowing foran optimum result to be obtained more often.

Optimising 2D Pose Representation: Improve Accuracy, Stability and Generalisability Within Unsupervised 2D-3D Human Pose Estimation