Research map / Video / Video generation
UniHand: A Unified Model for Diverse Controlled 4D Hand Motion Modeling
Zhihao Sun, Tong Wu, Ruirui Tu, Daoguo Dong, Zuxuan Wu
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 hand perceptron instead of average pooling of vision tokens (ablation) | UniHand · DexYCB (all) | PA-MPJPE (mm): 7.81 -> 4.08 | Improved |
| add hand perceptron instead of average pooling of vision tokens (ablation) | UniHand · HOT3D | PA-MPJPE (mm): 12.46 -> 4.76 | Improved |
| use pretrained vision encoder (ablation) | UniHand · DexYCB (all) | PA-MPJPE (mm): 6.52 -> 4.08 | Improved |
| use joint VAE condition encoders instead of an MLP (ablation) | UniHand · DexYCB (all) | PA-MPJPE (mm): 5.21 -> 4.08 | Improved |
| use 3D RoPE instead of 1D RoPE (ablation) | UniHand · DexYCB (all) | PA-MPJPE (mm): 4.65 -> 4.08 | Improved |
| add vision to 2D keypoint condition instead of 2D alone (ablation) | UniHand · DexYCB (all) | PA-MPJPE (mm): 4.75 -> 4.08 | Improved |
| add 3D keypoint condition to vision instead of 2D keypoints (ablation) | UniHand · DexYCB (all) | PA-MPJPE (mm): 4.08 -> 3.48 | Improved |
| use vision only instead of vision plus 2D keypoints (ablation) | UniHand · HOT3D | G-MPJPE (mm): 63.97 -> 53.49 G-MPJPE; PA-MPJPE 4.76 -> 4.52 | Mixed |
| UniHand canonical-space model vs HaWoR with explicit camera parameters | UniHand · HOT3D (world space) | G-MPJPE (mm): 47.35 -> 63.97 | Hurt |
Later papers that compared with this method
1 later accepted paper reports UniHand as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- StableHand: Quality-Aware Flow Matching for World-Space Dual-Hand Motion Estimation from Egocentric Video (NeurIPS 2026 · better on 2 of 2 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
- The Surprising Effectiveness of Video Diffusion Models for Hand Motion Reconstruction (NeurIPS 2026 · similarity 0.91)
- FoundHand: Large-Scale Domain-Specific Learning for Controllable Hand Image Generation (CVPR 2025 · similarity 0.90)
- StableHand: Quality-Aware Flow Matching for World-Space Dual-Hand Motion Estimation from Egocentric Video (NeurIPS 2026 · similarity 0.90)
- BOTH2Hands: Inferring 3D Hands from Both Text Prompts and Body Dynamics (CVPR 2024 · similarity 0.89)
- HandDiffuse: Generative Controllers for Two-Hand Interactions via Diffusion Models (AAAI 2025 · similarity 0.88)
- Dyn-HaMR: Recovering 4D Interacting Hand Motion from a Dynamic Camera (CVPR 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.
