Research map / Point clouds / Shape and surface completion
D-SCo: Dual-Stream Conditional Diffusion for Monocular Hand-Held Object Reconstruction
Bowen Fu, Gu Wang, Chenyangguang Zhang, Yan Di, Ziqin Huang, Zhiying Leng, Fabian Manhardt, Xiangyang Ji, Federico Tombari
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 projective mask loss (ablation) | D-SCo · ObMan | F-5: 0.57 -> 0.61 | Improved |
| use dual-stream denoiser instead of concatenating semantic and geometric embeddings (ablation) | D-SCo · ObMan | F-5: 0.54 -> 0.57 | Improved |
| add unified hand-object semantic embedding (ablation) | D-SCo · HO3D | F-5: 0.30 -> 0.34 | Improved |
| add hand articulation geometric embedding (ablation) | D-SCo · HO3D | F-5: 0.33 -> 0.34 | Improved |
| use GCN hand embedding instead of the proposed embeddings (ablation) | D-SCo · ObMan | F-5: 0.54 -> 0.52 | Hurt |
| use global PointNet hand embedding instead of unified semantic embedding (ablation) | D-SCo · HO3D | F-5: 0.34 -> 0.31 | Hurt |
| add hand-constrained centroid fixing (ablation) | D-SCo · ObMan | F-5: 0.44 -> 0.48 | Improved |
| add hand-constrained centroid prediction network (ablation) | D-SCo · ObMan | F-5: 0.32 -> 0.48 | Improved |
| give ground-truth object centroid at test without centroid fixing (ablation) | D-SCo without centroid fixing · ObMan | F-5: 0.44 -> 0.45 | No clear effect |
| select best of five diffusion samples (oracle) | D-SCo · HO3D | F-5: 0.41 -> 0.51 | Improved |
| zero-shot transfer from ObMan to HO3D versus DDF-HO | D-SCo · HO3D zero-shot | Chamfer Distance: F-5 0.27 vs 0.24; CD 1.19 vs 0.73 | Mixed |
Closest papers by meaning
- Learning Explicit Contact for Implicit Reconstruction of Hand-Held Objects from Monocular Images (AAAI 2024 · similarity 0.87)
- Coherent 3D Scene Diffusion From a Single RGB Image (NeurIPS 2024 · similarity 0.87)
- 6D-Diff: A Keypoint Diffusion Framework for 6D Object Pose Estimation (CVPR 2024 · similarity 0.87)
- GeneOH Diffusion: Towards Generalizable Hand-Object Interaction Denoising via Denoising Diffusion (ICLR 2024 · similarity 0.86)
- HandBooster: Boosting 3D Hand-Mesh Reconstruction by Conditional Synthesis and Sampling of Hand-Object Interactions (CVPR 2024 · similarity 0.86)
- Single-view Image to Novel-view Generation for Hand-Object Interactions (AAAI 2025 · 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.
