Research map / 3D vision / 3D generation
Diverse 3D Hand Gesture Prediction From Body Dynamics by Bilateral Hand Disentanglement
Xingqun Qi, Chen Liu, Muyi Sun, Lincheng Li, Changjie Fan, Xin Yu
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 |
|---|---|---|---|
| Use bilateral hand disentanglement with the complete stage-one method | 3D hand prediction from body dynamics · B2H and TED Hands | FHD: 0.973 -> 0.471; 0.877 -> 0.258 | Improved |
| Apply PSS diversification | Ours (stage one) · B2H | Diversity: 1.001$^{\pm 0.061}$ -> 1.324$^{\pm 0.066}$ | Improved |
| Apply PSS diversification | Ours (stage one) · TED Hands | Diversity: 4.018$^{\pm 0.190}$ -> 6.426$^{\pm 0.300}$ | Improved |
| Add the disentanglement loss (ablation) | Bilateral hand prediction model · B2H and TED Hands | FHD: 0.773 -> 0.682; 0.731 -> 0.595 | Improved |
| Add the perceptual loss (ablation) | Bilateral hand prediction model · B2H and TED Hands | FHD: 0.773 -> 0.657; 0.731 -> 0.699 | Improved |
| Add both disentanglement and perceptual losses (ablation) | Bilateral hand prediction model · B2H and TED Hands | FHD: 0.773 -> 0.640; 0.731 -> 0.495 | Improved |
| Add SRM (ablation) | Bilateral hand prediction model with both losses · B2H and TED Hands | FHD: 0.640 -> 0.580; 0.495 -> 0.393 | Improved |
| Add TMM after SRM (ablation) | Bilateral hand prediction model with both losses · B2H and TED Hands | FHD: 0.580 -> 0.471; 0.393 -> 0.258 | Improved |
| Place TMM before SRM rather than after it (ablation) | Bilateral hand prediction model · B2H and TED Hands | FHD: 0.471 -> 0.614; 0.258 -> 0.463 | Hurt |
| Share weights between the two hand transformers (ablation) | Bilateral hand prediction model · B2H and TED Hands | FHD: 0.471 -> 0.598; 0.258 -> 0.380 | Hurt |
Closest papers by meaning
- BOTH2Hands: Inferring 3D Hands from Both Text Prompts and Body Dynamics (CVPR 2024 · similarity 0.90)
- How Do I Do That? Synthesizing 3D Hand Motion and Contacts for Everyday Interactions (CVPR 2025 · similarity 0.89)
- Hand-Centric Motion Refinement for 3D Hand-Object Interaction via Hierarchical Spatial-Temporal Modeling (AAAI 2024 · similarity 0.89)
- Forecasting of 3D Whole-body Human Poses with Grasping Objects (CVPR 2024 · similarity 0.89)
- MEgoHand: Multimodal Egocentric Hand-Object Interaction Motion Generation (NeurIPS 2025 · similarity 0.88)
- HandDiffuse: Generative Controllers for Two-Hand Interactions via Diffusion Models (AAAI 2025 · similarity 0.88)
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.
