Cross-Modal Diffusion for Biomechanical Dynamical Systems Through Local Manifold Alignment

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Dey, Sharmita, Nair, Sarath Ravindran
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913740233900032
author Dey, Sharmita
Nair, Sarath Ravindran
author_facet Dey, Sharmita
Nair, Sarath Ravindran
contents We present a mutually aligned diffusion framework for cross-modal biomechanical motion generation, guided by a dynamical systems perspective. By treating each modality, e.g., observed joint angles ($X$) and ground reaction forces ($Y$), as complementary observations of a shared underlying locomotor dynamical system, our method aligns latent representations at each diffusion step, so that one modality can help denoise and disambiguate the other. Our alignment approach is motivated by the fact that local time windows of $X$ and $Y$ represent the same phase of an underlying dynamical system, thereby benefiting from a shared latent manifold. We introduce a simple local latent manifold alignment (LLMA) strategy that incorporates first-order and second-order alignment within the latent space for robust cross-modal biomechanical generation without bells and whistles. Through experiments on multimodal human biomechanics data, we show that aligning local latent dynamics across modalities improves generation fidelity and yields better representations.
format Preprint
id arxiv_https___arxiv_org_abs_2503_12214
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Cross-Modal Diffusion for Biomechanical Dynamical Systems Through Local Manifold Alignment
Dey, Sharmita
Nair, Sarath Ravindran
Machine Learning
Dynamical Systems
We present a mutually aligned diffusion framework for cross-modal biomechanical motion generation, guided by a dynamical systems perspective. By treating each modality, e.g., observed joint angles ($X$) and ground reaction forces ($Y$), as complementary observations of a shared underlying locomotor dynamical system, our method aligns latent representations at each diffusion step, so that one modality can help denoise and disambiguate the other. Our alignment approach is motivated by the fact that local time windows of $X$ and $Y$ represent the same phase of an underlying dynamical system, thereby benefiting from a shared latent manifold. We introduce a simple local latent manifold alignment (LLMA) strategy that incorporates first-order and second-order alignment within the latent space for robust cross-modal biomechanical generation without bells and whistles. Through experiments on multimodal human biomechanics data, we show that aligning local latent dynamics across modalities improves generation fidelity and yields better representations.
title Cross-Modal Diffusion for Biomechanical Dynamical Systems Through Local Manifold Alignment
topic Machine Learning
Dynamical Systems
url https://arxiv.org/abs/2503.12214