PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
March 29, 2026IEEE Transactions on Visualization and Computer Graphics0 citations

FCMD: Fine-Grained Text-Driven Cohesive Motion Generation With Diffusion Model

View Full Paper
SLShuai LiSWSiqi WangXZXinyu Zhang

Key Points

  • The research aims to generate continuous and expressive human motion from textual descriptions while ensuring coherence and realism.
  • Proposed a novel diffusion-based model called FCMD for motion generation.
  • Introduced Fine-grained Text Fusion for enhanced semantic consistency.
  • Implemented History Motion Guidance for accuracy across frames.
  • Used Smooth Stitching Sampling for seamless motion transitions.
  • Employed a large language model to refine the motion datasets.
  • FCMD demonstrates superior performance compared to state-of-the-art methods.
  • Achieved highly controllable and coherent motion sequences.
  • Showed improved semantic consistency in motion transitions.

Abstract

Generating continuous and expressive human motion from textual descriptions is a critical challenge in applications such as gaming and filmmaking. Existing methods often struggle to maintain global coherence, realistic frame continuity, and smooth transitions. To address these limitations, we propose FCMD, a novel diffusion-based model for generating cohesive motion sequences from fine-grained textual descriptions. FCMD introduces three key innovations: (1) Fine-grained Text Fusion, which integrates detailed textual cues with transitional narratives to enhance semantic consistency; (2) History Motion Guidance, ensuring motion accuracy and consistency across consecutive frames; and (3) Smooth Stitching Sampling, which leverages preceding and current motion information to achieve seamless transitions. Additionally, FCMD employs a large language model (LLM) to refine motion datasets by extracting fine-grained textual descriptions. Extensive experiments demonstrate that FCMD outperforms state-of-the-art methods in generating coherent, natural, and highly controllable motion sequences.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Li et al. (2026) studied this question.

synapsesocial.com/papers/69c8c195de0f0f753b39be8dhttps://doi.org/10.1109/tvcg.2026.3677594
Ask AI
Helpful
Bookmark
Share
View Full Paper