Privacy-Preserving Video Conferencing via Thermal-Generative Images

Sheng-Yang ChiuYu-Ting HuangChieh-Ting LinYu-Chee TsengJen-Jee ChenMeng-Hsuan TuBo-Chen TungYuJou Nieh

Sheng-Yang ChiuYu-Ting HuangChieh-Ting Lin ...+4 YuJou Nieh

Mar 2023

摘要

Due to the COVID-19 epidemic, video conferencing has evolved as a newparadigm of communication and teamwork. However, private and personalinformation can be easily leaked through cameras during video conferencing.This includes leakage of a person's appearance as well as the contents in thebackground. This paper proposes a novel way of using online low-resolutionthermal images as conditions to guide the synthesis of RGB images, bringing apromising solution for real-time video conferencing when privacy leakage is aconcern. SPADE-SR (Spatially-Adaptive De-normalization with Self Resampling), avariant of SPADE, is adopted to incorporate the spatial property of a thermalheatmap and the non-thermal property of a normal, privacy-free pre-recorded RGBimage provided in a form of latent code. We create a PAIR-LRT-Human (LRT =Low-Resolution Thermal) dataset to validate our claims. The result enables aconvenient way of video conferencing where users no longer need to groomthemselves and tidy up backgrounds for a short meeting. Additionally, it allowsa user to switch to a different appearance and background during a conference.

展开全部

图表提取

论文十问

参考文献

被引用

提问与回答

暂无人提供速读十问回答

被引用

笔记

问答

领域