EMA: Effort Metric Attention for Anatomical Effort-Guided Human Motion Diffusion

2026-05-23

AI Summary

Existing text-driven human motion generation methods struggle to precisely control motion intensity, often relying on vague adverbial descriptions and lacking quantitative modeling of dynamic attributes such as rhythm and force. This work proposes an intensity control framework based on Effort Metric Attention (EMA), which for the first time translates the Time and Weight components from Laban Movement Analysis into computable kinematic joint metrics. By integrating these numerical effort signals into a diffusion model via cross-attention mechanisms, the method enables region-level, fine-grained motion intensity control without requiring post-hoc optimization. Experimental and user studies demonstrate that the generated motions exhibit near-monotonic alignment between specified effort levels and dynamic characteristics, significantly outperforming existing approaches in both metric consistency and body-part-level controllability.

Abstract

Human motion diffusion models can synthesize action sequences from text, but controlling motion intensity remains challenging. Existing approaches rely on effort-related adverbs, which are ambiguous and fail to capture quantitative aspects such as pacing, often resulting in flat and monotonous dynamics. We propose an intensity-control framework based on Effort Metric Attention (EMA), a cross-attention module that conditions diffusion on numerical effort signals. Inspired by Laban Movement Analysis (LMA), the framework focuses on the Time and Weight effort factors. We approximate these factors using two kinematic metrics: peak joint positional change for pacing and collective joint positional change for motion amount. EMA enables fine-grained, region-wise control without costly post-hoc optimization. We introduce two evaluation tasks, metric-to-motion consistency and body-part-level effort modulation, to assess numerical fidelity and localized control. Experiments and a user study show near-monotonic alignment between specified effort levels, generated motion dynamics, and established LMA descriptors. These results indicate effective and interpretable control of effort dynamics in practice.

Problem

• human motion diffusion
• motion intensity control
• effort dynamics
• Laban Movement Analysis
• quantitative effort signals

Innovation

• Effort Metric Attention
• Human Motion Diffusion
• Laban Movement Analysis
• Intensity Control
• Kinematic Metrics