Research map / 3D vision / Face and shape
PaMIR: Parametric Model-Conditioned Implicit Representation for Image-Based Human Reconstruction
Zerong Zheng, Tao Yu, Yebin Liu, Qionghai Dai
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 depth-ambiguity-aware loss with predicted SMPL (ablation) | PaMIR · Twindom | P2S (cm): 1.81 -> 1.60 | Improved |
| train with predicted SMPL plus DAAL instead of ground-truth SMPL (ablation) | PaMIR · Twindom | P2S (cm): 1.71 -> 1.60 | Improved |
| add body reference optimization at inference (ablation) | PaMIR · Twindom | P2S (cm): 1.60 -> 1.52 | Improved |
| add body reference optimization (SMPL pose accuracy) (ablation) | PaMIR · BUFF | MPJPE (cm): 2.65 -> 2.49 | Improved |
| add body reference optimization (SMPL pose accuracy) (ablation) | PaMIR · Twindom | MPJPE (cm): 2.85 -> 2.74 | Improved |
| fuse features from more input views (1 to 4) (ablation) | PaMIR multi-image · multi-view normal reprojection | Normal reprojection error: 0.161 -> 0.135 | Improved |
| replace PIFu implicit function with SMPL-conditioned implicit function | PaMIR · BUFF | P2S (cm): 1.93 -> 1.52 | Improved |
| replace PIFu implicit function with SMPL-conditioned implicit function | PaMIR · Twindom | P2S (cm): 2.34 -> 1.80 | Improved |
| use ground-truth SMPL at test time instead of predicted | PaMIR · BUFF | P2S (cm): 1.52 -> 0.709 | 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.
- DeepHuman — DeepHuman: 3D Human Reconstruction From a Single Image (ICCV 2019) · better on 4 of 4 shared results
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
- Combining Implicit Function Learning and Parametric Models for 3D Human Reconstruction (ECCV 2020 · similarity 0.89)
- Multi-Person Implicit Reconstruction From a Single Image (CVPR 2021 · similarity 0.85)
- SHARP: Shape-Aware Reconstruction of People in Loose Clothing (IJCV 2022 · similarity 0.85)
- HISR: Hybrid Implicit Surface Representation for Photorealistic 3D Human Reconstruction (AAAI 2024 · similarity 0.85)
- HiLo: Detailed and Robust 3D Clothed Human Reconstruction with High-and Low-Frequency Information of Parametric Models (CVPR 2024 · similarity 0.83)
- PoseIRM: Enhance 3D Human Pose Estimation on Unseen Camera Settings via Invariant Risk Minimization (CVPR 2024 · similarity 0.83)
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.
