Research map / Video / Video generation
Think Then React: Towards Unconstrained Action-to-Reaction Motion Generation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes Think-Then-React (TTR), a framework for unconstrained action-to-reaction human motion generation that builds on a unified motion representation (details not in the provided abstract).
Problem
Generating human-like reactions to a given action sequence is hard because reactions cannot be directly predicted from actions without prompts, and there is no unified representation effectively encoding both action and reaction motion.
Approach
Frames action-to-reaction generation as a think-then-react process so that the reaction is produced after an intermediate reasoning step rather than directly from the raw action sequence; the abstract only partially states the mechanism.
Closest papers by meaning
- ReGenNet: Towards Human Action-Reaction Synthesis (CVPR 2024 · similarity 0.90)
- InterDreamer: Zero-Shot Text to 3D Dynamic Human-Object Interaction (NeurIPS 2024 · similarity 0.89)
- InterControl: Zero-shot Human Interaction Generation by Controlling Every Joint (NeurIPS 2024 · similarity 0.88)
- Ready-to-React: Online Reaction Policy for Two-Character Interaction Generation (ICLR 2025 · similarity 0.87)
- AMD: Autoregressive Motion Diffusion (AAAI 2024 · similarity 0.87)
- Enhanced Fine-Grained Motion Diffusion for Text-Driven Human Motion Synthesis (AAAI 2024 · 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.