AI Paper Atlas

Research map / 3D vision / Human pose estimation

TransPose: Keypoint Localization via Transformer

Sen Yang, Zhibin Quan, Mu Nie, Wankou Yang

ICCV 2021 · 3D vision › Human pose estimation · approach: Heatmap and coordinate pose encoding · 493 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 Transformer encoder on shallow ResNet backbone (TransPose-R-A4 vs SimpleBaseline-Res152)ResNet-S · COCO val2017AP: 72.0 -> 72.6Improved
add Transformer encoder on shallow HRNet backbone (TransPose-H-A6 vs HRNet-W48)HRNet-S · COCO val2017AP: 75.1 -> 75.8Improved
add position embedding (2D sine) to TransPose-R-A3 (ablation)TransPose-R-A3 · COCO val2017AP: 70.4 -> 71.7Improved
use learnable instead of no position embedding (ablation)TransPose-R-A3 · COCO val2017AP: 70.4 -> 70.9Improved
use fixed 2D sine instead of learnable position embedding (ablation)TransPose-R-A3 · COCO val2017AP: 70.9 -> 71.7Improved
add encoder layers from 2 to 4 in TransPose-R (ablation)TransPose-R · COCO val2017AP: 69.6 -> 72.6Improved
add encoder layers from 4 to 6 in TransPose-R (ablation)TransPose-R · COCO val2017AP: 72.6 -> 72.2Hurt
add encoder layers from 4 to 6 in TransPose-H (ablation)TransPose-H · COCO val2017AP: 75.3 -> 75.8Improved
increase embedding dimension d from 64 to 192 in TransPose-H (ablation)TransPose-H · COCO val2017AP: 75.1 -> 75.4Improved
COCO pre-training then fine-tuning on MPII instead of full-training (ablation)TransPose-R-A4 · MPII valMean@0.5: 89.3 -> 92.0Improved
COCO pre-training then fine-tuning on MPII instead of full-training (ablation)TransPose-H-A6 · MPII valMean@0.5: 90.3 -> 92.3Improved
COCO pre-training then fine-tuning on MPII instead of full-training (ablation)DARK-HRNet · MPII valMean@0.5: 90.6 -> 92.0Improved

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