AI Paper Atlas

Research map / 3D vision / Human pose estimation

Camera Distortion-Aware 3D Human Pose Estimation in Video With Optimization-Based Meta-Learning

Hanbyel Cho, Yooshin Cho, Jaemyung Yu, Junmo Kim

ICCV 2021 · 3D vision › Human pose estimation · approach: Monocular 2D-to-3D pose lifting · 22 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
add MAML with synthetic distorted tasks (ablation)Pavllo et al. base model · Human3.6M distorted d1/d2, Scenario 1, 9 framesMPJPE (mm): 84.2 / 79.6 -> 73.5 / 67.5Improved
add stratified sampling of distortion parameter (ablation)MAML with synthetic tasks · Human3.6M distorted d1/d2, Scenario 1, 9 framesMPJPE (mm): 73.5 / 67.5 -> 71.7 / 66.2Improved
add random distortion pretraining (ablation)MAML with stratified sampling · Human3.6M distorted d1/d2, Scenario 1, 9 framesMPJPE (mm): 71.7 / 66.2 -> 67.2 / 61.9Improved
add MAML with synthetic distorted tasks (ablation)Pavllo et al. base model · Human3.6M distorted d1/d2, Scenario 1, 9 framesPCKh@0.5 (%): 64.8 / 66.9 -> 72.0 / 74.5Improved
generate synthetic tasks from predicted 2D keypoints rather than ground-truth 3D joints (ablation)Proposed method · Human3.6M distorted, Scenario 1, heavyMPJPE (mm): 64.7 -> 62.0Improved
generate synthetic tasks from predicted 2D keypoints rather than ground-truth 3D joints (ablation)Proposed method · Human3.6M distorted, Scenario 2, heavyMPJPE (mm): 71.3 -> 66.1Improved
use proposed meta-learning method instead of baseline pose estimatorsPavllo et al. · Human3.6M distorted, Scenario 2, heavyMPJPE (mm): 114.1 -> 66.1Improved
increase input frame length from 3 to 27 (ablation)Proposed method · Human3.6M distorted, Scenario 1, all distortionsMPJPE (mm): 75.0 -> 57.6Improved
increase input frame length from 3 to 9 (ablation)Proposed method · Human3.6M distorted, Scenario 1, all distortionsMPJPE (mm): 75.0 -> 59.8Improved
apply proposed method with 27 frames instead of Pavllo 27fPavllo et al. · Human3.6M distorted, Scenario 1, all distortionsMPJPE (mm): 72.4 -> 57.6Improved
pretrain on random distortion without meta-learning (ablation)Random distortion pretraining · Human3.6M distortedaccuracy after adaptationHurt

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