AI Paper Atlas

Research map / 3D vision / Human pose estimation

MAMMA: Markerless Accurate Multi-person Motion Acquisition

Hanz Cuevas Velasquez, Anastasios Yiannakidis, Soyong Shin, Giorgio Becherini, Markus Höschle, Joachim Tesch, Taylor Obersat, Tsvetelina Alexiadis, Eni Halilaj, Michael J. Black

CVPR 2026 · 3D vision › Human pose estimation · approach: Physics and sensor-based motion capture

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
condition landmark network on SAM2 segmentation masks (ablation)MammaNet · Harmony4D two-person (IoU>0.5)2D landmark error (px): 31.96 -> 18.33Improved
condition landmark network on SAM2 masks (ablation)MammaNet · RICH single person2D landmark error (px): 8.55 -> 8.83No clear effect
condition on masks (ablation)MammaNet · CHI3D two-person2D landmark error (px): 6.22 -> 4.36Improved
use one learnable query per landmark instead of a single query for all landmarks (ablation)MammaNet vs CameraHMR decoder · MOYO2D landmark error (px): 12.53 -> 11.40Improved
add person-person contact optimization stage (ablation)MAMMA · Harmony4D, CHI3D, MammaEval-Dmean penetration depth (mm): 10.50 -> 8.46Improved
add contact optimization stage (ablation)MAMMA · 3D fitting MPJPE benchmarksMPJPE: Harmony4D 45.26 -> 45.35; CHI3D 38.01 -> 37.96No clear effect
optimize floor contact (ablation)MAMMA · 3D fittingreconstruction accuracyNo clear effect
markerless MAMMA versus Vicon+MoSh++ marker-based pipelineMAMMA · MammaEval-Extraheld-out marker error (mm): 21.619mm vs 22.481mmNo clear effect
match people across views via symmetric epipolar affinity of landmarksMAMMA correspondence · Hi4D, Harmony4D, CHI3D, MammaEval-Didentity matching accuracy: 100% accuracyImproved

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