Research map / 3D vision / Human pose estimation
SAME: Spatial-Aware Multimodal Egocentric Human Pose Estimation
Yurong Fu, Peng Dai, Yu Zhang, Feng Yiqiang, Yang Zhang, Haoqian Wang
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 spatial-aware multi-view fusion (depth-guided 3D PE) (ablation) | SAME · EMHI Protocol2 | MPJPE: 40.7 -> 38.0 | Improved |
| add modality-aware multi-modal fusion in place of vanilla Transformer decoder (ablation) | SAME · EMHI Protocol2 | MPJPE: 38.8 -> 38.0 | Improved |
| add temporal Transformer network in global frame (ablation) | SAME · EMHI Protocol2 | Ground: 24.5 -> 4.7 | Improved |
| add temporal Transformer network in global frame (ablation) | SAME · EMHI Protocol2 | MPJPE: 39.0 -> 38.0 | Improved |
| add ground-penetration loss (ablation) | SAME · EMHI Protocol2 | Ground: 8.1 -> 4.7 | Improved |
| add ground-penetration loss (ablation) | SAME · EMHI Protocol2 | MPJPE: 38.5 -> 38.0 | Improved |
| add IMU to stereo-only input (ablation) | SAME · EMHI Protocol2 | MPJPE: 62.4 -> 38.0 | Improved |
| add stereo images to IMU-only input (ablation) | SAME · EMHI Protocol2 | MPJPE: 52.1 -> 38.0 | Improved |
| add right camera view to IMU plus left camera (ablation) | SAME · EMHI Protocol2 | MPJPE: 41.0 -> 38.0 | Improved |
| mid-level fusion of IMU and stereo versus MEPoser concatenation | SAME · EMHI Protocol2 | MPJPE: 47.6 -> 38.0 | Improved |
Closest papers by meaning
- EgoPoseFormer: A Simple Baseline for Stereo Egocentric 3D Human Pose Estimation (ECCV 2024 · similarity 0.89)
- EMHI: A Multimodal Egocentric Human Motion Dataset with HMD and Body-Worn IMUs (AAAI 2025 · similarity 0.89)
- Scene-Aware Egocentric 3D Human Pose Estimation (CVPR 2023 · similarity 0.89)
- SelfPose: 3D Egocentric Pose Estimation From a Headset Mounted Camera (TPAMI 2020 · similarity 0.88)
- 3D Human Pose Perception from Egocentric Stereo Videos (CVPR 2024 · similarity 0.87)
- FRAME: Floor-aligned Representation for Avatar Motion from Egocentric Video (CVPR 2025 · similarity 0.87)
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.
