Research map / 3D vision / 3D generation
Generating Diverse and Natural 3D Human Motions From Text
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use text2length sampling rather than real motion lengths | Text2motion generation · HumanML3D | MultiModality: 2.090±.083 -> 2.219±.074 | Helped |
| Use text2length sampling rather than real motion lengths | Text2motion generation · HumanML3D | FID: 1.067±.002 -> 1.087±.021 | Hurt |
| Use text2length sampling rather than real motion lengths | Text2motion generation · KIT-ML | MultiModality: 1.482±.065 -> 2.052±.107 | Helped |
| Use text2length sampling rather than real motion lengths | Text2motion generation · KIT-ML | FID: 2.770±.109 -> 3.022±.107 | Hurt |
| Add motion snippet codes (ablation) | Text2motion generation · HumanML3D | R Precision Top 1: 0.370±.002 -> 0.455±.003 | Helped |
| Add motion snippet codes (ablation) | Text2motion generation · HumanML3D | FID: 1.200±.027 -> 1.087±.021 | Helped |
| Add local word attention (ablation) | Text2motion generation · HumanML3D | R Precision Top 1: 0.396±.002 -> 0.455±.003 | Helped |
| Add local word attention (ablation) | Text2motion generation · HumanML3D | FID: 1.833±.032 -> 1.087±.021 | Helped |
| Add part-of-speech tags (ablation) | Text2motion generation · HumanML3D | R Precision Top 1: 0.443±.003 -> 0.455±.003 | Helped |
| Add part-of-speech tags (ablation) | Text2motion generation · HumanML3D | FID: 1.157±.016 -> 1.087±.021 | Helped |
| Add time-to-arrival positional encoding (ablation) | Text2motion generation · HumanML3D | R Precision Top 1: 0.444±.005 -> 0.455±.003 | Helped |
| Add time-to-arrival positional encoding (ablation) | Text2motion generation · HumanML3D | FID: 1.229±.020 -> 1.087±.021 | Helped |
Closest papers by meaning
- SnapMoGen: Human Motion Generation from Expressive Texts (NeurIPS 2025 · similarity 0.92)
- Generating Human Motion in 3D Scenes from Text Descriptions (CVPR 2024 · similarity 0.91)
- AMD: Autoregressive Motion Diffusion (AAAI 2024 · similarity 0.90)
- MMM: Generative Masked Motion Model (CVPR 2024 · similarity 0.89)
- InterDreamer: Zero-Shot Text to 3D Dynamic Human-Object Interaction (NeurIPS 2024 · similarity 0.88)
- Enhancing Motion in Text-to-Video Generation with Decomposed Encoding and Conditioning (NeurIPS 2024 · similarity 0.88)
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.