Research map / 3D vision / Human pose estimation
MVHumanNet++: A Large-Scale Dataset of Multi-View Daily Dressing Human Captures With Richer Annotations for 3D Human Digitization
Chenghong Li, Hongjie Liao, Yihao Zhi, Xihe Yang, Zhengwentai Sun, Jiahao Chang, Shuguang Robert Cui, Xiaoguang Han
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 |
|---|---|---|---|
| train on 8 views instead of 1 view | FR-Head · MVHumanNet++ action recognition | top-1 accuracy (%): 30.25 -> 78.19 | Improved |
| train on 8 views instead of 1 view | InfoGCN · MVHumanNet++ action recognition | top-1 accuracy (%): 25.23 -> 76.55 | Improved |
| scale NeRF training outfits from 100 to 5000 | IBRNet · MVHumanNet++ generalizable NeRF | PSNR: 26.05 -> 29.00 | Improved |
| scale NeRF training outfits from 100 to 5000 | GPNeRF · MVHumanNet++ generalizable NeRF | PSNR: 23.27 -> 24.69 | Improved |
| pretrain on MVHumanNet++ then finetune on HuMMan | IBRNet · HuMMan test | PSNR: 28.06 -> 29.46 | Improved |
| pretrain on MVHumanNet++ then finetune on HuMMan | GPNeRF · HuMMan test | PSNR: 20.95 -> 21.89 | Improved |
| use MVHumanNet++ pretrained model without finetuning | IBRNet · HuMMan test | PSNR: 28.06 -> 27.48 | Hurt |
| use improved SMPLX annotations from MVHumanNet++ (ablation) | Animatable Gaussians · MVHumanNet++ per-subject 3DGS | PSNR: 27.259 -> 28.593 | Improved |
| scale LaRa training outfits from 100 to 5000 | LaRa · MVHumanNet++ generalizable 3DGS | PSNR: 21.431 -> 22.441 | Improved |
| scale EVA-Gaussian training outfits from 100 to 5000 | EVA-Gaussian · MVHumanNet++ generalizable 3DGS | PSNR: 26.984 -> 28.544 | Improved |
| scale training subjects from 3000 to 5500 | StyleGAN2 · MVHumanNet++ generation | FID: 14.05 -> 7.08 | Improved |
| scale training subjects from 3000 to 5500 | GET3D · MVHumanNet++ generation | FID: 41.54 -> 25.12 | Improved |
Closest papers by meaning
- MVHumanNet: A Large-scale Dataset of Multi-view Daily Dressing Human Captures (CVPR 2024 · similarity 0.98)
- PKU-DyMVHumans: A Multi-View Video Benchmark for High-Fidelity Dynamic Human Modeling (CVPR 2024 · similarity 0.88)
- 4D-DRESS: A 4D Dataset of Real-World Human Clothing With Semantic Annotations (CVPR 2024 · similarity 0.87)
- OpenHumanVid: A Large-Scale High-Quality Dataset for Enhancing Human-Centric Video Generation (CVPR 2025 · similarity 0.87)
- HumanVid: Demystifying Training Data for Camera-controllable Human Image Animation (NeurIPS 2024 · similarity 0.86)
- BEDLAM2.0: Synthetic humans and cameras in motion (NeurIPS 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.
