Research map / 3D vision / Human pose estimation
Neural Marionette: Unsupervised Learning of Motion Skeleton and Latent Dynamics from Volumetric Video
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use the learned skeleton instead of P2S | Neural Marionette skeleton extraction · AIST++, HanCo, and Panda | SC-score: 0.755 -> 0.804; 0.847 -> 0.944; 0.937 -> 0.954 | Helped |
| Use the learned skeleton instead of the hand-labeled GT skeleton | 4D tracking · HanCo and Panda | Chamfer distance (×10^4): 2.12 -> 1.97; 1.77 -> 1.71 | Helped |
| Use the learned skeleton instead of the hand-labeled GT skeleton | 4D tracking · AIST++ | Chamfer distance (×10^4): 3.11 -> 3.42 | Hurt |
| Encode local joint rotations with forward kinematics rather than directly regress node positions | Neural Marionette motion dynamics · HanCo motion interpolation, Average | Motion Chamfer Distance: 2.98 -> 2.64 | Helped |
| Encode local joint rotations with forward kinematics rather than directly regress node positions | Neural Marionette motion dynamics · D-FAUST motion interpolation, Average | Motion Chamfer Distance: 2.96 -> 2.88 | Helped |
| Add the volume fitting loss L_vol (ablation) | Neural Marionette skeleton module · AIST++ | SC-score: 0.603 -> 0.804 | Helped |
| Add the volume fitting loss L_vol (ablation) | Neural Marionette skeleton module · HanCo | SC-score: 0.890 -> 0.944 | Helped |
| Add the volume fitting loss L_vol (ablation) | Neural Marionette skeleton module · AIST++ | Chamfer distance (×10^4): 16.7 -> 3.42 | Helped |
| Add the volume fitting loss L_vol (ablation) | Neural Marionette skeleton module · HanCo | Chamfer distance (×10^4): 2.30 -> 1.97 | Helped |
| Add the graph trajectory loss L_traj (ablation) | Neural Marionette affinity estimation · AIST++ and HanCo | skeleton bone alignment to rigid parts | Helped |
Closest papers by meaning
- SAMoR: Motion Modelling for Articulated Objects of Any Skeleton and Topology (NeurIPS 2026 · similarity 0.89)
- Towards Natural and Accurate Future Motion Prediction of Humans and Animals (CVPR 2019 · similarity 0.88)
- Learning Implicit Representation for Reconstructing Articulated Objects (ICLR 2024 · similarity 0.88)
- Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action Recognition (CVPR 2025 · similarity 0.87)
- Neuron: Learning Context-Aware Evolving Representations for Zero-Shot Skeleton Action Recognition (CVPR 2025 · similarity 0.87)
- Heterogeneous Skeleton-Based Action Representation Learning (CVPR 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.