Research map / Reinforcement learning / Visual and representation RL
ConTrack: Constrained Hand Motion Tracking with Adaptive Trade-off Control
Yutong Liang, Quanquan Peng, Rizhao Qiu, Xiaolong 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 adaptive task-style mixing instead of fixed task weight (ablation) | ConTrack · hand-object tracking | progress: 0.764 -> 0.899 | Improved |
| add adaptive task-style mixing instead of fixed task weight (ablation) | ConTrack · hand-object tracking | contact F1: 0.679 -> 0.784 | Improved |
| add adaptive mixing instead of fixed 1:1 mixing (ablation) | ConTrack · hand-object tracking | progress: 0.868 -> 0.899 | Improved |
| add adaptive mixing instead of fixed 1:1 mixing (ablation) | ConTrack · hand-object tracking | contact F1: 0.701 -> 0.784 | Improved |
| use fixed task-only weight vs adaptive on object rotation error (ablation) | ConTrack · hand-object tracking | object rotation error (rad): 0.250 -> 0.272 | Hurt |
| add adaptive reset library instead of reset from start (ablation) | ConTrack · hand-object tracking | progress: 0.700 -> 0.899 | Improved |
| add adaptive reset library instead of uniform mid-clip reset (ablation) | ConTrack · hand-object tracking | progress: 0.727 -> 0.899 | Improved |
| add adaptive reset library instead of reset from start (ablation) | ConTrack · hand-object tracking | object position error (m): 0.039 -> 0.026 | Improved |
| add contact reward (ablation) | ConTrack · hand-object tracking | contact F1: 0.699 -> 0.784 | Improved |
| add contact reward (ablation) | ConTrack · hand-object tracking | progress: 0.861 -> 0.899 | Improved |
| add contact distance reward (ablation) | ConTrack · hand-object tracking | contact F1: 0.753 -> 0.784 | Improved |
| add contact distance reward (ablation) | ConTrack · hand-object tracking | finger error (rad): 0.149 -> 0.163 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- ManipTrans — ManipTrans: Efficient Dexterous Bimanual Manipulation Transfer via Residual Learning (CVPR 2025) · better on 4 of 6 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
- DexTrack: Towards Generalizable Neural Tracking Control for Dexterous Manipulation from Human References (ICLR 2025 · similarity 0.89)
- ComPose: When to Trust Hands for Object Pose Tracking (NeurIPS 2026 · similarity 0.87)
- Scaffolding Dexterous Manipulation with Vision-Language Models (NeurIPS 2025 · similarity 0.86)
- SkiP: When to Skip and When to Refine for Efficient Robot Manipulation (NeurIPS 2026 · similarity 0.85)
- Implicit Drifting Policy: One-Step Action Generation via Conditional Expert Geometry (NeurIPS 2026 · similarity 0.85)
- CoDA: Coordinated Diffusion Noise Optimization for Whole-Body Manipulation of Articulated Objects (NeurIPS 2025 · similarity 0.85)
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.
