Research map / 3D vision / Human pose estimation
NeRMo: Learning Implicit Neural Representations for 3D Human Motion Prediction
Dong Wei, Huaijiang Sun, Xiaoning Sun, Shengxiang Hu
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 Fourier features for temporal coordinate encoding (ablation) | NeRMo · CMU-MoCap | MPJPE 80ms: 14.83 -> 8.12 | Improved |
| add codebook-coordinate attention on top of Fourier features (ablation) | NeRMo · CMU-MoCap | MPJPE 400ms: 38.39 -> 37.15 | Improved |
| add codebook-coordinate attention on top of Fourier features (ablation) | NeRMo · CMU-MoCap | MPJPE 1000ms: 83.81 -> 80.46 | Improved |
| use spatial-temporal (joint-code) representation instead of temporal-only (ablation) | NeRMo · CMU-MoCap | MPJPE | Improved |
| add mask-aware spatial attention (ablation) | NeRMo · complete and incomplete observation | MPJPE | Improved |
| use meta-optimization (ablation) | NeRMo · CMU-MoCap | MPJPE | Improved |
| use a continuous INR model instead of discrete historical-value models with 20% missing values | NeRMo · Human3.6M 20% missing | MPJPE reduction vs best baseline: 10.7%, 7.9%, 6.0% and 4.1% at 80, 160, 320, 400ms | Improved |
| use NeRMo for long-horizon prediction on complete observation | NeRMo · Human3.6M average | MPJPE 400ms: 58.3 -> 59.1 vs SPGSN | No clear effect |
| repair missing values first (R+) before predicting | LTD, DMGNN · Human3.6M 20% missing | MPJPE: 24.5 -> 19.9 (LTD walking 80ms) | Improved |
Later papers that compared with this method
1 later accepted paper reports NeRMo as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- ALIEN: Implicit Neural Representations for Human Motion Prediction under Arbitrary Latency (CVPR 2025 · better on 4 of 4 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
- Investigating Pose Representations and Motion Contexts Modeling for 3D Motion Prediction (TPAMI 2022 · similarity 0.90)
- Towards Natural and Accurate Future Motion Prediction of Humans and Animals (CVPR 2019 · similarity 0.89)
- NeRM: Learning Neural Representations for High-Framerate Human Motion Synthesis (ICLR 2024 · similarity 0.89)
- ALIEN: Implicit Neural Representations for Human Motion Prediction under Arbitrary Latency (CVPR 2025 · similarity 0.89)
- Existence Is Chaos: Enhancing 3D Human Motion Prediction with Uncertainty Consideration (AAAI 2024 · similarity 0.87)
- FutureHuman3D: Forecasting Complex Long-Term 3D Human Behavior from Video Observations (CVPR 2024 · 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.
