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 current state-of-the-art in monocular 3D human pose estimation is heavilyinfluenced by weakly supervised methods. These allow 2D labels to be used tolearn effective 3D human pose recovery either directly from images or via2D-to-3D pose uplifting. In this paper we present a detailed analysis of themost commonly used simplified projection models, which relate the estimated 3Dpose representation to 2D labels: normalized perspective and weak perspectiveprojections. Specifically, we derive theoretical lower bound errors for thoseprojection models under the commonly used mean per-joint position error(MPJPE). Additionally, we show how the normalized perspective projection can bereplaced to avoid this guaranteed minimal error. We evaluate the derived lowerbounds on the most commonly used 3D human pose estimation benchmark datasets.Our results show that both projection models lead to an inherent minimal errorbetween 19.3mm and 54.7mm, even after alignment in position and scale. This isa considerable share when comparing with recent state-of-the-art results. Ourpaper thus establishes a theoretical baseline that shows the importance ofsuitable projection models in weakly supervised 3D human pose estimation.

Error Bounds of Projection Models in Weakly Supervised 3D Human Pose Estimation