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 present LInKs, a novel unsupervised learning method to recover 3D humanposes from 2D kinematic skeletons obtained from a single image, even whenocclusions are present. Our approach follows a unique two-step process, whichinvolves first lifting the occluded 2D pose to the 3D domain, followed byfilling in the occluded parts using the partially reconstructed 3D coordinates.This lift-then-fill approach leads to significantly more accurate resultscompared to models that complete the pose in 2D space alone. Additionally, weimprove the stability and likelihood estimation of normalising flows through acustom sampling function replacing PCA dimensionality reduction previously usedin prior work. Furthermore, we are the first to investigate if different partsof the 2D kinematic skeleton can be lifted independently which we find byitself reduces the error of current lifting approaches. We attribute this tothe reduction of long-range keypoint correlations. In our detailed evaluation,we quantify the error under various realistic occlusion scenarios, showcasingthe versatility and applicability of our model. Our results consistentlydemonstrate the superiority of handling all types of occlusions in 3D spacewhen compared to others that complete the pose in 2D space. Our approach alsoexhibits consistent accuracy in scenarios without occlusion, as evidenced by a7.9% reduction in reconstruction error compared to prior works on the Human3.6Mdataset. Furthermore, our method excels in accurately retrieving complete 3Dposes even in the presence of occlusions, making it highly applicable insituations where complete 2D pose information is unavailable.

LInKs "Lifting Independent Keypoints" -- Partial Pose Lifting for Occlusion Handling with Improved Accuracy in 2D-3D Human Pose Estimation