AI Paper Atlas

Research map / 3D vision / Human pose estimation

Sequential 3D Human Pose and Shape Estimation From Point Clouds

Kangkan Wang, Jin Xie, Guofeng Zhang, Lei Liu, Jian Yang

CVPR 2020 · 3D vision › Human pose estimation · approach: Category-level pose estimation · 54 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
Predict mesh vertex coordinates directly with spatial-temporal mesh attention convolutionSequential point-cloud body reconstruction · SURREALReconstruction error (mm): 49.5 -> 18.2Improved
Predict mesh vertex coordinates directly with spatial-temporal mesh attention convolutionSequential point-cloud body reconstruction · Human3.6MReconstruction error (mm): 54.3 -> 21.4Improved
Predict mesh vertex coordinates directly with spatial-temporal mesh attention convolutionSequential point-cloud body reconstruction · DFAUSTReconstruction error (mm): 52.2 -> 19.7Improved
Add spatial mesh attention convolution (SMAC) (ablation)Proposed sequential mesh reconstruction · SURREAL, Human3.6M, DFAUSTReconstruction error (mm): 45.3 -> 18.2; 49.1 -> 21.4; 47.6 -> 19.7Improved
Use learned spatial attention instead of neighborhood averaging (ablation)Proposed sequential mesh reconstruction · SURREAL, Human3.6M, DFAUSTReconstruction error (mm): 25.7 -> 18.2; 26.5 -> 21.4; 25.9 -> 19.7Improved
Add temporal mesh attention convolution (TMAC) (ablation)Proposed sequential mesh reconstruction · SURREAL, Human3.6M, DFAUSTReconstruction error (mm): 21.2 -> 18.2; 22.9 -> 21.4; 22.2 -> 19.7Improved
Regress SMPL parameters rather than mesh vertices through the proposed mesh-convolution framework (ablation)Point-cloud body reconstruction · SURREAL, Human3.6M, DFAUSTReconstruction error (mm): 18.2 -> 65.8; 21.4 -> 68.4; 19.7 -> 65.3Hurt
Increase the sampled input point count from 2,500 to 5,000 (ablation)Proposed sequential mesh reconstruction · SURREAL, Human3.6M, DFAUSTReconstruction error (mm): 18.2 -> 18.0; 21.4 -> 20.8; 19.7 -> 19.3Improved
Increase the sampled input point count from 2,500 to 7,500 (ablation)Proposed sequential mesh reconstruction · SURREAL, Human3.6M, DFAUSTReconstruction error (mm): 18.2 -> 17.9; 21.4 -> 20.7; 19.7 -> 19.1Improved
Fine-tune with weak supervision using point correspondences and surface regularizersProposed model on real clothed-body point clouds · Real detailed data3D model fit to input point cloudsImproved

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