Research map / 3D vision / Face and shape
Identity-Preserving Human Reconstruction from a Single Image via 3D Token Inference
Yanqi Bao, Jiaxiang Shang, Yang Gao, Yingchun Liu, Jing Huo, Jing Liao
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 feature projection to voxel tokens (ablation) | IPRM · Synthetic Data | PSNR (input view): 17.01 -> 28.90 | Improved |
| use voxel resolution 128 instead of 64 (ablation) | IPRM · Synthetic Data | PSNR (input view): 27.36 -> 28.90 | Improved |
| add visibility-based cross-attention instead of self-attention (ablation) | IPRM · Synthetic Data | PSNR (input view): 27.87 -> 28.90 | Improved |
| add adaptive 3D Human Feature (ablation) | IPRM · Synthetic Data | PSNR (input view): 28.54 -> 28.90 | Improved |
| add 2D image condition ID adapter (ablation) | IPRM · Synthetic Data | PSNR (input view): 25.80 -> 26.42 | Improved |
| replace 2D ID adapter with 3D ID Adapter (ablation) | IPRM · Synthetic Data | PSNR (input view): 26.42 -> 28.90 | Improved |
| add 3D ID Adapter over no condition (ablation) | IPRM · Synthetic Data | PSNR (input view): 25.80 -> 28.90 | Improved |
| add back-view input | IPRM · THuman2.1 | PSNR: 26.74 -> 27.31 | Improved |
| use all views as input | IPRM · THuman2.1 | PSNR: 26.74 -> 27.69 | Improved |
| replace LHM with IPRM sparse voxel token inference | LHM-0.5B · Synthetic Data | PSNR: 25.18 -> 27.04 | Improved |
| use IPRM instead of per-case optimized PSHuman for mesh | PSHuman · THuman2.1 (mesh) | Chamfer Distance: 0.4399 -> 0.4458 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 4 on most shared results.
- PSHuman — PSHuman: Photorealistic Single-image 3D Human Reconstruction using Cross-Scale Multiview Diffusion and Explicit Remeshing (CVPR 2025) · better on 2 of 4 shared results
- SIFu — SIFU: Side-view Conditioned Implicit Function for Real-world Usable Clothed Human Reconstruction (CVPR 2024) · better on 4 of 4 shared results
- SiTH — SiTH: Single-view Textured Human Reconstruction with Image-Conditioned Diffusion (CVPR 2024) · better on 2 of 2 shared results
- DreamGaussian — DreamGaussian: Generative Gaussian Splatting for Efficient 3D Content Creation (ICLR 2024) · 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
- TokenHMR: Advancing Human Mesh Recovery with a Tokenized Pose Representation (CVPR 2024 · similarity 0.88)
- SiTH: Single-view Textured Human Reconstruction with Image-Conditioned Diffusion (CVPR 2024 · similarity 0.87)
- SyncHuman: Synchronizing 2D and 3D Generative Models for Single-view Human Reconstruction (NeurIPS 2025 · similarity 0.86)
- GeneMAN: Generalizable Single-Image 3D Human Reconstruction from Multi-Source Human Data (NeurIPS 2025 · similarity 0.86)
- PSHuman: Photorealistic Single-image 3D Human Reconstruction using Cross-Scale Multiview Diffusion and Explicit Remeshing (CVPR 2025 · similarity 0.86)
- D^3-Human: Dynamic Disentangled Digital Human from Monocular Video (CVPR 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.
