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 task of estimating a physically plausible articulated humanmotion from monocular video. Existing approaches that do not consider physicsoften produce temporally inconsistent output with motion artifacts, whilestate-of-the-art physics-based approaches have either been shown to work onlyin controlled laboratory conditions or consider simplified body-ground contactlimited to feet. This paper explores how these shortcomings can be addressed bydirectly incorporating a fully-featured physics engine into the pose estimationprocess. Given an uncontrolled, real-world scene as input, our approachestimates the ground-plane location and the dimensions of the physical bodymodel. It then recovers the physical motion by performing trajectoryoptimization. The advantage of our formulation is that it readily generalizesto a variety of scenes that might have diverse ground properties and supportsany form of self-contact and contact between the articulated body and scenegeometry. We show that our approach achieves competitive results with respectto existing physics-based methods on the Human3.6M benchmark, while beingdirectly applicable without re-training to more complex dynamic motions fromthe AIST benchmark and to uncontrolled internet videos.

Trajectory Optimization for Physics-Based Reconstruction of 3d Human Pose from Monocular Video