AI Paper Atlas

Research map / 3D vision / Human pose estimation

Joint Optimization for 4D Human-Scene Reconstruction in the Wild

Zhizheng Liu, Joe Lin, Wayne Wu, Bolei Zhou

ICLR 2026 · 3D vision › Human pose estimation · approach: Physics and sensor-based motion capture · 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
joint optimization of camera, human and scene with contact losses instead of sequential optimization (same HMR2.0+DROID-SLAM init)SynCHMR* · SLOPER4Djitter: 123.9 -> 7.6Improved
same, foot floatingSynCHMR* · SLOPER4Dfoot floating rate (%): 9.0 -> 3.3Improved
upgrade initialization to VIMO + MASt3R with joint optimizationJOSH · SLOPER4DWA-MPJPE (mm): 233.2 -> 120.0Improved
jointly optimize with HMR2.0 init instead of SLAHMRSLAHMR · EMDB W-MPJPEW-MPJPE (mm): 776.1 -> 372.3Improved
joint optimization on top of VIMO (vs TRAM)TRAM · EMDBW-MPJPE (mm): 222.4 -> 174.7Improved
add joint optimization to MASt3R scene reconstructionMASt3R · SLOPER4Dchamfer distance: 57.0% lessImproved
add contact scene loss L_c1 (ablation)JOSH_3 · SLOPER4Dfoot floating rate (%): 92.9 -> 2.9Improved
add contact scene loss L_c1 (ablation)JOSH_3 · SLOPER4DATE (m): 22.47 -> 3.21Improved
also optimize human parameters (not only scene and camera) (ablation)JOSH_3 · SLOPER4DW-MPJPE (mm): 486.4 -> 438.3Improved
add contact static loss L_c2 (ablation)JOSH_3 · SLOPER4Dfoot sliding (mm): 68.2 -> 28.2Improved
optimize focal length jointly (ablation)JOSH_3 (estimated intrinsics) · SLOPER4DW-MPJPE (mm): 1220.7 -> 1053.5Improved
train JOSH3R on JOSH-labeled web videos instead of EMDB ground truthJOSH3R · EMDBWA-MPJPE: 59.2% lowerImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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