Geometry-Aware Losses for Structure-Preserving Text-to-Sign Language Generation

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Wu, Zetian, Zhou, Tianshuo, Lee, Stefan, Huang, Liang
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866915517363650560
author Wu, Zetian
Zhou, Tianshuo
Lee, Stefan
Huang, Liang
author_facet Wu, Zetian
Zhou, Tianshuo
Lee, Stefan
Huang, Liang
contents Sign language translation from text to video plays a crucial role in enabling effective communication for Deaf and hard--of--hearing individuals. A major challenge lies in generating accurate and natural body poses and movements that faithfully convey intended meanings. Prior methods often neglect the anatomical constraints and coordination patterns of human skeletal motion, resulting in rigid or biomechanically implausible outputs. To address this, we propose a novel approach that explicitly models the relationships among skeletal joints--including shoulders, arms, and hands--by incorporating geometric constraints on joint positions, bone lengths, and movement dynamics. During training, we introduce a parent-relative reweighting mechanism to enhance finger flexibility and reduce motion stiffness. Additionally, bone-pose losses and bone-length constraints enforce anatomically consistent structures. Our method narrows the performance gap between the previous best and the ground-truth oracle by 56.51%, and further reduces discrepancies in bone length and movement variance by 18.76% and 5.48%, respectively, demonstrating significant gains in anatomical realism and motion naturalness.
format Preprint
id arxiv_https___arxiv_org_abs_2509_23011
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Geometry-Aware Losses for Structure-Preserving Text-to-Sign Language Generation
Wu, Zetian
Zhou, Tianshuo
Lee, Stefan
Huang, Liang
Computer Vision and Pattern Recognition
Computation and Language
Sign language translation from text to video plays a crucial role in enabling effective communication for Deaf and hard--of--hearing individuals. A major challenge lies in generating accurate and natural body poses and movements that faithfully convey intended meanings. Prior methods often neglect the anatomical constraints and coordination patterns of human skeletal motion, resulting in rigid or biomechanically implausible outputs. To address this, we propose a novel approach that explicitly models the relationships among skeletal joints--including shoulders, arms, and hands--by incorporating geometric constraints on joint positions, bone lengths, and movement dynamics. During training, we introduce a parent-relative reweighting mechanism to enhance finger flexibility and reduce motion stiffness. Additionally, bone-pose losses and bone-length constraints enforce anatomically consistent structures. Our method narrows the performance gap between the previous best and the ground-truth oracle by 56.51%, and further reduces discrepancies in bone length and movement variance by 18.76% and 5.48%, respectively, demonstrating significant gains in anatomical realism and motion naturalness.
title Geometry-Aware Losses for Structure-Preserving Text-to-Sign Language Generation
topic Computer Vision and Pattern Recognition
Computation and Language
url https://arxiv.org/abs/2509.23011