Research map / 3D vision / Human pose estimation
Nymeria: A Massive Collection of Egocentric Multi-modal Human Motion in the Wild
Lingni Ma, Yuting Ye, Rowan Postyeni, Alexander J Gamino, Vijay Baiyya, Luis Pesqueira, Kevin M Bailey, David Soriano Fosas, Fangzhou Hong, Vladimir Guzov, Yifeng Jiang, Hyo Jin Kim and 5 more
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 AvatarPoser with real SLAM device poses instead of synthetic mocked-up poses (ablation) | AvatarPoser · Nymeria 3-point tracking | MPJPE mean (cm): 7.31 -> 7.97 | Hurt |
| train and test AvatarPoser on Nymeria instead of AMASS | AvatarPoser · 3-point tracking | MPJPE mean (cm): 4.20 -> 7.31 | Hurt |
| use 3-point BoDiffusion instead of 1-point EgoEgo for motion synthesis | motion synthesis · Nymeria motion synthesis | MPJPE mean (cm): 13.22 -> 7.98 | Improved |
| add product quantization (PQ=2) to motion VQ-VAE (ablation) | motion VQ-VAE · Nymeria motion reconstruction | MPJPE (mm): 51.60 -> 39.63 | Improved |
| increase codebook size from 2048 to 16384 and reduce latent dim from 512 to 64 (ablation) | motion VQ-VAE · Nymeria motion reconstruction | MPJPE (mm): 39.63 -> 34.49 | Improved |
| use MotionGPT (T5 language prior) instead of TM2T for motion-to-text | motion-to-text on Nymeria · Nymeria motion-to-text | CIDEr: 20.85 -> 37.27 | Improved |
Closest papers by meaning
- EgoSim: An Egocentric Multi-view Simulator and Real Dataset for Body-worn Cameras during Motion and Activity (NeurIPS 2024 · similarity 0.87)
- EgoVid-5M: A Large-Scale Video-Action Dataset for Egocentric Videos Generation (NeurIPS 2025 · similarity 0.86)
- Harmony4D: A Video Dataset for In-The-Wild Close Human Interactions (NeurIPS 2024 · similarity 0.86)
- EgoLM: Multi-Modal Language Model of Egocentric Motions (CVPR 2025 · similarity 0.86)
- EgoGen: An Egocentric Synthetic Data Generator (CVPR 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.
