Research map / 3D vision / Face and shape
SesaHand: Enhancing 3D Hand Reconstruction via Controllable Generation with Semantic and Structural Alignment
Zhuoran Zhao, Xianghao Kong, Linlin Yang, Zheng WEI, Pan Hui, Anyi Rao
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 human behavior semantics extraction (CoT) to caption prompts (ablation) | hand image generation · MSCOCO | FID: 21.04 -> 19.83 | Improved |
| use human behavior semantics instead of raw VLM captions as prompt (ablation) | ControlNet-based hand generation · Prompt influence analysis (100 samples) | hand detector confidence: 86% -> 97% | Improved |
| add hierarchical structural fusion (ablation) | SesaHand · MSCOCO | FID | Improved |
| add hand structure attention enhancement (ablation) | SesaHand · MSCOCO | FID | Improved |
| set cross-attention bias alpha to 2.0 versus other values (ablation) | hand structure attention enhancement · MSCOCO | generation performance | Improved |
| use SesaHand generated images instead of AttentionHand's | InterWild · HIC | MPVPE: 14.74 -> 14.70 | Improved |
| use SesaHand generated images instead of AttentionHand's | InterWild · ReIH | MPVPE: 13.95 -> 13.01 | Improved |
| use SesaHand generated images instead of AttentionHand's | DIR · ReIH | MPVPE: 19.91 -> 19.21 | Improved |
| fine-tune with SesaHand images | InterWild · ReIH | MRRPE: 22.38 -> 22.17 | Mixed |
| fine-tune with SesaHand images | DIR · ReIH | RRVE: 29.41 -> 27.42 | Mixed |
| use SesaHand instead of AttentionHand for generation | hand image generation · MSCOCO | KID: 0.00301 -> 0.00553 | Hurt |
| use SesaHand instead of AttentionHand for generation | hand image generation · MSCOCO | FID-H: 27.09 -> 17.77 | Improved |
Closest papers by meaning
- FoundHand: Large-Scale Domain-Specific Learning for Controllable Hand Image Generation (CVPR 2025 · similarity 0.89)
- RHanDS: Refining Malformed Hands for Generated Images with Decoupled Structure and Style Guidance (AAAI 2025 · similarity 0.89)
- HandBooster: Boosting 3D Hand-Mesh Reconstruction by Conditional Synthesis and Sampling of Hand-Object Interactions (CVPR 2024 · similarity 0.88)
- Fine-Grained Multi-View Hand Reconstruction Using Inverse Rendering (AAAI 2024 · similarity 0.88)
- SRHand: Super-Resolving Hand Images and 3D Shapes via View/Pose-aware Neural Image Representations and Explicit Meshes (NeurIPS 2025 · similarity 0.88)
- HanDiffuser: Text-to-Image Generation With Realistic Hand Appearances (CVPR 2024 · similarity 0.88)
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.
