Research map / 3D vision / 3D generation
Capturing Hands in Action Using Discriminative Salient Points and Physics Simulation
Dimitrios Tzionas, Luca Ballan, Abhilash Srikantha, Pablo Aponte, Marc Pollefeys, Jüergen Gall
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 collision term (gamma_c=10) to local optimization (ablation) | LO · RGB-D hand-hand sequences (11) | mean 2D error (px): 5.53 -> 5.17 | Improved |
| add salient point detector to local optimization (ablation) | LO · RGB-D hand-hand sequences (11) | mean 2D error (px): 5.53 -> 5.34 | Improved |
| add both collision and salient terms (ablation) | LO · RGB-D hand-hand sequences (11) | mean 2D error (px): 5.53 -> 5.12 | Improved |
| fit both model-to-data and data-to-model instead of one only (ablation) | LO · RGB-D hand-hand sequences (11) | failed sequences: 11/11 -> 0/11 | Improved |
| use distance-field collision term instead of vertex-correspondence repulsion (ablation) | LO+S+C · RGB-D hand-hand sequences (11) | mean 2D error (px): 6.66 -> 5.12 | Improved |
| small collision weights gamma_c between 0 and 3 (ablation) | LO+S+C · RGB-D hand-hand sequences (11) | mean 2D error (px): 5.34 -> 5.57 | Hurt |
| collision weight gamma_c from 0 to 10 (ablation) | LO+S+C · RGB-D hand-hand sequences (11) severe collision | mean 2D error (px): 5.90 -> 5.55 | Improved |
| collision weight gamma_c from 0 to 10 (ablation) | LO+S+C · RGB-D hand-hand sequences (11) some collision | mean 2D error (px): 3.99 -> 3.99 | No clear effect |
| use point-to-plane instead of point-to-point at 10 iterations (ablation) | LO · RGB-D hand-hand sequences (11) | mean 2D error (px): 5.25 -> 5.12 | Improved |
| use point-to-plane instead of point-to-point at 30 iterations (ablation) | LO · RGB-D hand-hand sequences (11) | mean 2D error (px): 4.91 -> 5.05 | Hurt |
| add salient points and collision to local optimization (ablation) | LO · Dexter (excluding random) | mean error (mm): 17.86 -> 14.26 | Improved |
| add salient point detector (accurate detector) (ablation) | LO · Dexter (excluding random) | mean error (mm): 17.86 -> 15.51 | Improved |
Closest papers by meaning
- ComPose: When to Trust Hands for Object Pose Tracking (NeurIPS 2026 · similarity 0.89)
- HOLD: Category-agnostic 3D Reconstruction of Interacting Hands and Objects from Video (CVPR 2024 · similarity 0.88)
- HandDiffuse: Generative Controllers for Two-Hand Interactions via Diffusion Models (AAAI 2025 · similarity 0.87)
- Hand-held Object Reconstruction from RGB Video with Dynamic Interaction (CVPR 2025 · similarity 0.87)
- Hand-Centric Motion Refinement for 3D Hand-Object Interaction via Hierarchical Spatial-Temporal Modeling (AAAI 2024 · similarity 0.87)
- DiffGrasp: Whole-Body Grasping Synthesis Guided by Object Motion Using a Diffusion Model (AAAI 2025 · 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.
