Research map / Video / Video generation
HVG-3D: Bridging Real and Simulation Domains for 3D-Conditional Hand-Object Interaction Video Synthesis
Mingjin Chen, Junhao Chen, Zhaoxin Fan, Yujian Lee, Zichen Dang, Lili Wang, Yawen Cui, Lap-Pui Chau, Yi 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 3D point cloud condition (ablation) | HVG-3D · TASTE-Rob ablation | PSNR: 18.44 -> 24.15 | Improved |
| add 3D point cloud condition (ablation) | HVG-3D · TASTE-Rob ablation | SSIM: 0.37 -> 0.81 | Improved |
| add 3D point cloud condition (ablation) | HVG-3D · TASTE-Rob ablation | LPIPS: 0.5957 -> 0.193 | Improved |
| add 3D tracking video condition (ablation) | HVG-3D · TASTE-Rob ablation | PSNR: 22.76 -> 24.15 | Improved |
| add 3D tracking video condition (ablation) | HVG-3D · TASTE-Rob ablation | SSIM: 0.75 -> 0.81 | Improved |
| add 3D tracking video condition (ablation) | HVG-3D · TASTE-Rob ablation | LPIPS: 0.2054 -> 0.193 | Improved |
| add mask diffusion loss (ablation) | HVG-3D · TASTE-Rob ablation | PSNR: 22.09 -> 24.15 | Improved |
| add mask diffusion loss (ablation) | HVG-3D · TASTE-Rob ablation | SSIM: 0.80 -> 0.81 | Improved |
| add mask diffusion loss (ablation) | HVG-3D · TASTE-Rob ablation | ST-SSIM: 0.96 -> 0.97 | Improved |
| add 3D point cloud condition (ablation) | HVG-3D · TASTE-Rob ablation | ST-SSIM: 0.91 -> 0.97 | Improved |
| compare HVG-3D with DaS in the hand-object region instead of full frame | HVG-3D vs DaS · TASTE-Rob full frame | L1: 7.77 vs 9.50 | Hurt |
| evaluate in the hand-object masked region instead of full frame | HVG-3D · TASTE-Rob | FVD: 13.8 -> 9.6 | Improved |
Closest papers by meaning
- SViMo: Synchronized Diffusion for Video and Motion Generation in Hand-object Interaction Scenarios (NeurIPS 2025 · similarity 0.90)
- HOIDiffusion: Generating Realistic 3D Hand-Object Interaction Data (CVPR 2024 · similarity 0.87)
- HandBooster: Boosting 3D Hand-Mesh Reconstruction by Conditional Synthesis and Sampling of Hand-Object Interactions (CVPR 2024 · similarity 0.87)
- Re-HOLD: Video Hand Object Interaction Reenactment via adaptive Layout-instructed Diffusion Model (CVPR 2025 · similarity 0.87)
- HO-Cap: A Capture System and Dataset for 3D Reconstruction and Pose Tracking of Hand-Object Interaction (NeurIPS 2025 · similarity 0.86)
- HOI-Swap: Swapping Objects in Videos with Hand-Object Interaction Awareness (NeurIPS 2024 · similarity 0.86)
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.
