AI Paper Atlas

Research map / 3D vision / Human pose estimation

Maximum-Margin Structured Learning with Deep Networks for 3D Human Pose Estimation

Sijin Li, Weichen Zhang, Antoni B. Chan

IJCV 2016 · 3D vision › Human pose estimation · approach: Monocular 2D-to-3D pose lifting · 187 citations (OpenAlex)

Read the paperCheck 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
average the top-500 scoring poses instead of taking the max pose (ablation)StructNet · Human3.6M test (S9,S11), all actionsMPJPE (mm): 135.63 -> 121.46Improved
average top-20 poses instead of max (ablation)StructNet · Human3.6M test (S9,S11), all actionsMPJPE (mm): 135.63 -> 128.11Improved
increase averaged poses from 500 to 1500 (ablation)StructNet · Human3.6M test (S9,S11), all actionsMPJPE (mm): 121.46 -> 123.04No clear effect
apply APF to the averaged pose to get a valid pose (ablation)StructNet · Human3.6M test (S9,S11), all actionsMPJPE (mm): 121.46 -> 121.31No clear effect
add auxiliary pose prediction task (lambda=1) (ablation)StructNet · Human3.6M test (S9,S11), all actionsMPJPE (mm): 145.74 -> 135.63Improved
use 2-layer pose embedding instead of raw joint coordinates (ablation)StructNet · Human3.6M test, WalkingMPJPE (mm): 114.24 -> 83.64Improved
use 1-layer learned embedding instead of fixed random weights (ablation)StructNet · Human3.6M test, WalkingMPJPE (mm): 87.94 -> 86.48Improved
use 2-layer instead of 1-layer pose embedding (ablation)StructNet · Human3.6M test, WalkingMPJPE (mm): 86.48 -> 83.64Improved
use StructNet-Avg(500)-APF over DconvMP-HMLStructNet · Human3.6M test (S9,S11), all actionsMPJPE (mm): 133.54 -> 121.31Improved
use StructNet-Avg(500) over LinKDE on hidden test setStructNet · Human3.6M online hidden testMPJPE (mm): 133.81 -> 122.03Improved
use score-function poses instead of auxiliary regression posesStructNet · Human3.6M test (S9,S11), all actionsMPJPE (mm): 133.79 -> 121.46Improved

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