Research map / 3D vision / Human pose estimation
Novel-View Synthesis and Pose Estimation for Hand-Object Interaction from Sparse Views
Wentian Qu, Zhaopeng Cui, Yinda Zhang, Chenyu Meng, Cuixia Ma, Xiaoming Deng, Hongan 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 online joint model fitting with render and pose losses (ablation) | hand-object rendering · HandObject rendering | PSNR: 22.07 -> 22.17 | Improved |
| add interaction loss (SDF penetration and contact) (ablation) | online fitting · HandObject rendering | PSNR: 22.17 -> 22.25 | Improved |
| add interaction loss (ablation) | online fitting · Synthetic DexYCB master-chef-can | penetration depth (mm): 9.65 -> 6.09 | Improved |
| add interaction loss (ablation) | online fitting · HandObject (bean-can, box, cup, meat-can) | hand MPJPE: bean-can 7.72 -> 8.07 | Mixed |
| add smooth loss (ablation) | video fitting · HandObject | Acc-V: 10.07 -> 6.35 | Improved |
| add stable contact loss (ablation) | video fitting with smooth loss · HandObject | PCI (%): 36.37 -> 51.18 | Improved |
| add stable contact loss (ablation) | video fitting with smooth loss · Synthetic DexYCB | Acc-J: 6.78 -> 7.51 | Hurt |
| use SDF-based pre-built hand model instead of A-NeRF density hand model (ablation) | hand-object fitting · HandObject (8 views) | object AD: 20.94 -> 15.95 | Improved |
| fit with color loss on neural models instead of untextured mesh fitting (ablation) | pose refinement · HandObject (8 views) | hand MPJPE: 20.60 -> 10.80 | Improved |
| refine LT hand pose with online model fitting | LT · HandObject (8 views) | hand MPJPE: 9.66 -> 9.09 | Improved |
| pre-built hand and object models as priors for sparse-view rendering | novel view synthesis · HandObject (3 views) | PSNR: 18.70 -> 22.20 vs InfoNeRF | Improved |
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.
- InfoNeRF — InfoNeRF: Ray Entropy Minimization for Few-Shot Neural Volume Rendering (CVPR 2022) · better on 1 of 1 shared result
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
- Single-view Image to Novel-view Generation for Hand-Object Interactions (AAAI 2025 · similarity 0.90)
- EasyHOI: Unleashing the Power of Large Models for Reconstructing Hand-Object Interactions in the Wild (CVPR 2025 · similarity 0.88)
- How Do I Do That? Synthesizing 3D Hand Motion and Contacts for Everyday Interactions (CVPR 2025 · similarity 0.88)
- ComPose: When to Trust Hands for Object Pose Tracking (NeurIPS 2026 · similarity 0.88)
- Hand-Centric Motion Refinement for 3D Hand-Object Interaction via Hierarchical Spatial-Temporal Modeling (AAAI 2024 · similarity 0.87)
- MEgoHand: Multimodal Egocentric Hand-Object Interaction Motion Generation (NeurIPS 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.
