AI Paper Atlas

Research map / 3D vision / Human pose estimation

Unsupervised Human Pose Estimation Through Transforming Shape Templates

Luca Schmidtke, Athanasios Vlontzos, Simon Ellershaw, Anna Lukens, Tomoki Arichi, Bernhard Kainz

CVPR 2021 · 3D vision › Human pose estimation · approach: Heatmap and coordinate pose encoding · 38 citations (OpenAlex)

Check your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use a part-based template without paired or unpaired pose labelsOurs · Human3.6M test%-MSE normalized by image size: 2.73 -> 3.31Hurt
Use a part-based template without unpaired pose labels instead of Jakab's methodInfant Dataset pose estimation · Infant Dataset test%-MSE normalized by image size: 8.98 -> 4.86Improved
Use the proposed 2D pose estimates as input to the 3D lifting network3D lifting · Synthetic Infants3D joint position error (mm): 59.7 -> 44.7Improved
Add boundary loss (ablation)Template pose estimator · Infant Dataset (10-epoch ablation)mean joint distance (% of image size): 65.99 -> 6.54Improved
Add anchor loss (ablation)Template pose estimator · Infant Dataset (10-epoch ablation)mean joint distance (% of image size): 12.42 -> 6.54Improved
Add both anchor and boundary losses (ablation)Template pose estimator · Infant Dataset (10-epoch ablation)mean joint distance (% of image size): 446.47 -> 6.54Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy