Diffusion-Based 3D Sign Language Motion Anonymization: A Feasibility Study on Balancing Identity Confusion and Semantic Preservation

Zixuan Dai, Shinji Sako


Abstract
Sign language motions contain individual-specific kinematic features. As the engineering applications of sign language become more widespread, privacy protection of sign language data has emerged as a new challenge. This paper proposes a diffusion model-based approach for sign language motion anonymization. The proposed framework combines conditional diffusion processes with adversarial training to transform identity features while preserving semantic information. For the design and preliminary validation of the proposed model, we conduct a proof-of-concept experiment using a subset of 22 signers from the ASL100 dataset of WLASL, which demonstrates the feasibility of the proposed approach for sign language anonymization.
Anthology ID:
2026.signlang-1.10
Volume:
Proceedings of the LREC 2026 12th Workshop on the Representation and Processing of Sign Languages: Language in Motion
Month:
May
Year:
2026
Address:
Palma, Mallorca (Spain)
Editors:
Eleni Efthimiou, Stavroula-Evita Fotinea, Thomas Hanke, Julie A. Hochgesang, Johanna Mesch, Marc Schulder
Venues:
SignLang | WS
SIG:
Publisher:
ELRA Language Resources Association (ELRA)
Note:
Pages:
93–99
Language:
External URL:
https://lrec.elra.info/lrec2026-ws-signlang-10
DOI:
10.63317/4nrz6i8vbmt2
Bibkey:
Cite (ACL):
Zixuan Dai and Shinji Sako. 2026. Diffusion-Based 3D Sign Language Motion Anonymization: A Feasibility Study on Balancing Identity Confusion and Semantic Preservation. In Proceedings of the LREC 2026 12th Workshop on the Representation and Processing of Sign Languages: Language in Motion, pages 93–99, Palma, Mallorca (Spain). ELRA Language Resources Association (ELRA).
Cite (Informal):
Diffusion-Based 3D Sign Language Motion Anonymization: A Feasibility Study on Balancing Identity Confusion and Semantic Preservation (Dai & Sako, SignLang 2026)
Copy Citation: