PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
October 16, 2025Open Access

E-React: Towards Emotionally Controlled Synthesis of Human Reactions

View Full Paper
Ask AI
Bookmark
Share

Authors

CZChen ZhuBHBuzhen HuangZWZhi‐Zheng Wu

Discussion

Loading...

Member takes

Overview

Novel approach generates diverse human reactions from emotional cues, highlighting the role of emotion in motion generation.

Key Points

  • Our model generates realistic reactions based on emotional conditions, showcasing significant improvements over existing methods.
  • Using a semi-supervised framework, we learn emotion representation effectively from limited motion data to enhance reaction synthesis.
  • The actor-reactor diffusion model captures both spatial interaction and emotional response for more natural human-like reactions.
  • Experimental results confirm that our approach outperforms traditional frameworks in generating emotionally responsive motions.

Cite This Study

Zhu et al. (2025) studied this question.

synapsesocial.com/papers/68f10ecee6a12fd042899a7chttps://doi.org/10.48550/arxiv.2508.06093
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1MoReact: Generating Reactive Motion from Textual Descriptions2025
  2. 2HERO: Human Reaction Generation from Videos2025
  3. 3ReGenNet: Towards Human Action-Reaction Synthesis2024
  4. 4Real-time and Controllable Reactive Motion Synthesis via Intention Guidance2025
  5. 5MARRS: Masked Autoregressive Unit-based Reaction Synthesis2026