Research map / Video / Video generation
Event-T2M: Event-level Conditioning for Complex Text-to-Motion Synthesis
Seong-Eun Hong, Jaeyoung Seon, JuYeong Hwang, JongHwan Shin, HyeongYeop Kang
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 event-level conditioning (>=4 events) | Event-T2M vs MoMask · HumanML3D-E >=4 | FID: 0.418 -> 0.265 | Improved |
| add event-level conditioning on standard test | Event-T2M vs MoGenTS · HumanML3D | FID: 0.033 -> 0.056 | Hurt |
| event-level conditioning on KIT-ML | Event-T2M vs MoGenTS · KIT-ML | R-Precision Top-1: 0.445 vs 0.439 | No clear effect |
| event-level conditioning vs MARDM on KIT-ML | Event-T2M vs MARDM-SiT · KIT-ML | R-Precision Top-1: 0.387 -> 0.379 | Hurt |
| event-level instead of token-level cross-attention (TMR) (ablation) | Event-T2M · HumanML3D-E >=4 | FID: 0.355 -> 0.265 | Improved |
| event-level instead of token-level (CLIP, >=4 events) (ablation) | Event-T2M · HumanML3D-E >=4 | R-Precision Top-1: 0.399 -> 0.374 | Hurt |
| replace CLIP with TMR text encoder (ablation) | Event-T2M · HumanML3D-E >=4 | R-Precision Top-1: 0.374 -> 0.466 | Improved |
| event-level conditioning on MModality | Event-T2M · HumanML3D | MModality: 2.799 (MDM) -> 0.949 | Hurt |
| increase event count from 2 to 4 | Event-T2M · HumanML3D-E | FID: 0.079 -> 0.265 | Hurt |
| event-aware LLM segmentation vs verb-aware | Event definition · user study 1 | naturalness rating | Improved |
Closest papers by meaning
- AMD: Autoregressive Motion Diffusion (AAAI 2024 · similarity 0.89)
- AToM: Aligning Text-to-Motion Model at Event-Level with GPT-4Vision Reward (CVPR 2025 · similarity 0.88)
- SnapMoGen: Human Motion Generation from Expressive Texts (NeurIPS 2025 · similarity 0.88)
- OMG: Towards Open-vocabulary Motion Generation via Mixture of Controllers (CVPR 2024 · similarity 0.88)
- Motion Prompting: Controlling Video Generation with Motion Trajectories (CVPR 2025 · similarity 0.87)
- DartControl: A Diffusion-Based Autoregressive Motion Model for Real-Time Text-Driven Motion Control (ICLR 2025 · similarity 0.87)
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.
