Research map / 3D vision / Human pose estimation
VQ-HPS: Human Pose and Shape Estimation in a Vector-Quantized Latent Space
Guénolé Fiche, Simon Leglaive, Xavier Alameda-Pineda, Antonio Agudo, Francesc Moreno
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 |
|---|---|---|---|
| predict discrete latent mesh indices instead of SMPL parameters (ablation) | VQ-HPS · 3DPW (trained on 3DPW) | PVE (mm): 199.8 -> 176.6 | Improved |
| train with cross-entropy on latent indices instead of a 3D vertex loss (ablation) | VQ-HPS · 3DPW (trained on 3DPW) | PVE (mm): 220.3 -> 176.6 | Improved |
| add reprojection loss (ablation) | VQ-HPS · 3DPW (trained on 3DPW) | PVE (mm): 183.9 -> 176.6 | Improved |
| use VQ-HPS instead of FastMETRO-S with scarce data (3DPW) | non-parametric HPSE · 3DPW (trained on 3DPW) | PVE (mm): 176.3 -> 163.9 | Improved |
| use VQ-HPS instead of FastMETRO-S when trained on COCO | EMDB (trained on COCO) · EMDB (trained on COCO) | PVE (mm): 143.1 -> 152.7 | Hurt |
| use VQ-HPS vs PARE with large-scale data (ResNet-50) | 3DPW (large-scale) · 3DPW (large-scale, no finetune) | PVE (mm): 99.7 -> 93.6 | Improved |
| add Bedlam to TokenHMR training data | TokenHMR · 3DPW | PVE (mm): 88.1 -> 84.6 | Improved |
Later papers that compared with this method
2 later accepted papers report VQ-HPS as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- MEGA: Masked Generative Autoencoder for Human Mesh Recovery (CVPR 2025 · better on 6 of 6 shared results)
- GenHMR: Generative Human Mesh Recovery (AAAI 2025 · better on 2 of 2 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- CLIFF — CLIFF: Carrying Location Information in Full Frames into Human Pose and Shape Estimation (ECCV 2022) · better on 2 of 2 shared results
- TokenHMR — TokenHMR: Advancing Human Mesh Recovery with a Tokenized Pose Representation (CVPR 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
- Convolutional Mesh Regression for Single-Image Human Shape Reconstruction (CVPR 2019 · similarity 0.90)
- Learning to Estimate 3D Human Pose and Shape From a Single Color Image (CVPR 2018 · similarity 0.90)
- PoseExaminer: Automated Testing of Out-of-Distribution Robustness in Human Pose and Shape Estimation (CVPR 2023 · similarity 0.88)
- Accurate 3D Body Shape Regression Using Metric and Semantic Attributes (CVPR 2022 · similarity 0.88)
- Neural Localizer Fields for Continuous 3D Human Pose and Shape Estimation (NeurIPS 2024 · similarity 0.87)
- $\text{Di}^2\text{Pose}$: Discrete Diffusion Model for Occluded 3D Human Pose Estimation (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.
