AI Paper Atlas

Research map / 3D vision / Human pose estimation

Geometric Neural Distance Fields for Learning Human Motion Priors

Zhengdi Yu, Simone Foti, Linguang Zhang, Amy Zhao, Cem Keskin, Stefanos Zafeiriou, Tolga Birdal

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
add transition and acceleration priors on top of pose prior (NRMF vs NRDF-t) (ablation)NRMF · AMASS motion denoisingPos. Err. All (mm): 22.8 -> 16.4Improved
add transition prior T-NRDF to SMPLer-X refinement (ablation)SMPLer-X + NRDF · 3DPW 2D fittingAcc Err (m/s^2): 24.31 -> 9.89Improved
add acceleration prior A-NRDF to SMPLer-X refinement (ablation)SMPLer-X + NRDF · 3DPW 2D fittingAcc Err (m/s^2): 24.31 -> 6.73Improved
combine pose, transition and acceleration priors (NRMF full) versus pose prior only (ablation)SMPLer-X refinement · 3DPW 2D fittingMPJPE (mm): 71.88 -> 66.13Improved
optimize SMPLer-X with no prior (ablation)SMPLer-X · 3DPW 2D fittingMPJPE (mm): 82.65 -> 84.67Hurt
use Riemannian-geometry NRDF-based transition prior instead of Euclidean PoseNDF (ablation)NRDF/PoseNDF pose prior · AMASS motion denoisingPos. Err. All (mm): 18.8 -> 16.7Improved
conditioned product-space NRMF instead of decoupled Motion-NDF (ablation)NRMF · AMASS motion denoisingAcc Err: 2.64 -> 2.25Improved
use pose prior only (-t) for partial observations (ablation)NRDF-t · AMASS partial 3D observationsPos. Err. Vis vs Occ (mm): Vis 4.2 but Occ 194.5Mixed
NRMF versus RoHM on partial observationsNRMF · AMASS partial 3D observationsPos. Err. All (mm): 51.3 -> 46.3Improved
NRMF versus RoHM on PROX RGB frequency of floor penetration distance (RGB)NRMF · PROX RGBDist: 14.31 -> 16.21Hurt
NRMF versus HuMoR on motion generationNRMF · Motion generationFID_m: 9.147 -> 5.317Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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