Research map / 3D vision / Human pose estimation
A Structure-aware and Motion-adaptive Framework for 3D Human Pose Estimation with Mamba
Ye Lu, Jie Wang, Jianjun Gao, Rui Gong, Chen Cai, Kim-Hui Yap
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add SSD module to stacked-block baseline (ablation) | SAMA-B · Human3.6M | MPJPE (mm): reduction of 0.6 mm in MPJPE from 39.9 | Improved |
| add SSI on top of SSD (ablation) | SAMA-B · Human3.6M | MPJPE (mm): 39.9 -> 38.4 | Improved |
| add MSM on top of SSI (ablation) | SAMA-B · Human3.6M | MPJPE (mm): 38.4 -> 37.4 | Improved |
| prepend SSI and MSM (ablation) | MixSTE · Human3.6M | MPJPE (mm): reduction of 0.6 mm | Improved |
| prepend SSI and MSM (ablation) | MotionBERT · Human3.6M | MPJPE (mm): reduction of 1.2 mm | Improved |
| prepend SSI and MSM (ablation) | MotionAGFormer · Human3.6M | MPJPE (mm): reduction of 0.9 mm; reaches 37.5 | Improved |
| replace scanning strategies (bi-directional, global-local) with SSI on vanilla scan (ablation) | SAMA-B · Human3.6M | MPJPE (mm): best MPJPE of 37.4 mm with lower computational cost | Improved |
| add timescale from adjacent-frame motion (point-wise convolution or linear) (ablation) | SAMA-B · Human3.6M | MPJPE (mm) | Improved |
| scale from SAMA-L T=243 to T=351 input frames | SAMA-L · Human3.6M | MPJPE P1 (mm): 36.9 -> 36.5 | Improved |
| SAMA-L versus PoseMamba-X | SAMA-L · Human3.6M (2D ground truth) | MPJPE (mm): 14.8 -> 11.4 | Improved |
| SAMA versus PoseMamba | SAMA · MPI-INF-3DHP | MPJPE (mm): 14.5 -> 14.4 | Improved |
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.
- KTPFormer — KTPFormer: Kinematics and Trajectory Prior Knowledge-Enhanced Transformer for 3D Human Pose Estimation (CVPR 2024) · better on 1 of 1 shared result
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
- PoseMamba: Monocular 3D Human Pose Estimation with Bidirectional Global-Local Spatio-Temporal State Space Model (AAAI 2025 · similarity 0.88)
- Sequential Joint Dependency Aware Human Pose Estimation with State Space Model (AAAI 2025 · similarity 0.87)
- SAMoR: Motion Modelling for Articulated Objects of Any Skeleton and Topology (NeurIPS 2026 · similarity 0.87)
- MV-SSM: Multi-View State Space Modeling for 3D Human Pose Estimation (CVPR 2025 · similarity 0.86)
- Optimal-state Dynamics Estimation for Physics-based Human Motion Capture from Videos (NeurIPS 2024 · similarity 0.86)
- Spatial-Mamba: Effective Visual State Space Models via Structure-Aware State Fusion (ICLR 2025 · similarity 0.86)
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.
