Diffusion Path Alignment for Long-Range Motion Generation and Domain Transitions

TL;DR

This paper introduces a diffusion-based inference optimization method for generating coherent, long-range human motion transitions across different semantic domains, enhancing applications like dance choreography.

Contribution

It presents the first general framework for controlled long-range human motion generation with explicit transition modeling using diffusion models.

Findings

Optimizing the control-energy objective improves transition fidelity.

The method yields temporally coherent long-range motion sequences.

Applicable to diverse semantic motion domains.

Abstract

Long-range human movement generation remains a central challenge in computer vision and graphics. Generating coherent transitions across semantically distinct motion domains remains largely unexplored. This capability is particularly important for applications such as dance choreography, where movements must fluidly transition across diverse stylistic and semantic motifs. We propose a simple and effective inference-time optimization framework inspired by diffusion-based stochastic optimal control. Specifically, a control-energy objective that explicitly regularizes the transition trajectories of a pretrained diffusion model. We show that optimizing this objective at inference time yields transitions with fidelity and temporal coherence. This is the first work to provide a general framework for controlled long-range human motion generation with explicit transition modeling.