arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2410.13675cs.CVcs.CL
收藏

基于姿态的手语外观迁移

Pose-Based Sign Language Appearance Transfer

University of Zurich(苏黎世大学)

Amit Moryossef, Gerard Sant, Zifan Jiang

AI总结:

提出一种在手语骨骼姿态中迁移打手势者外观并保留手语内容的方法,在掩盖身份改进渲染的同时揭示了隐私保护与手语识别性能之间的权衡。

摘要:

我们提出了一种在手语骨骼姿态中迁移打手势者外观同时保留手语内容的方法。利用估计的姿态,我们将一名打手势者的外观迁移给另一名,保持自然的动作和过渡。该方法在掩盖身份的同时,改进了基于姿态的渲染和手语拼接。我们的实验表明,虽然该方法降低了打手势者识别准确率,但也略微损害了手语识别性能,凸显了隐私与实用性之间的权衡。我们的代码可在 https://github.com/sign-language-processing/pose-anonymization 获取。

英文摘要:

We introduce a method for transferring the signer's appearance in sign language skeletal poses while preserving the sign content. Using estimated poses, we transfer the appearance of one signer to another, maintaining natural movements and transitions. This approach improves pose-based rendering and sign stitching while obfuscating identity. Our experiments show that while the method reduces signer identification accuracy, it slightly harms sign recognition performance, highlighting a tradeoff between privacy and utility. Our code is available at https://github.com/sign-language-processing/pose-anonymization.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑