Detailed Human Avatars from Monocular Video
Detailed Human Avatars from Monocular Video
复制标题
DOI：
10.1109/3dv.2018.00022
复制
发表时间：
2018-08
期刊：
影响因子：
--
通讯作者：
Thiemo Alldieck;M. Magnor;Weipeng Xu;C. Theobalt;Gerard Pons-Moll
中科院分区：
文献类型：
--
作者：
Thiemo Alldieck;M. Magnor;Weipeng Xu;C. Theobalt;Gerard Pons-Moll
提出了一种基于单目视频的高保留细节的人物头像创建方法。一个参数化的身体模型被细化和优化，以最大限度地像从各个方面展示他们的视频对象。我们的虚拟形象有自然的脸、发型、有褶皱的衣服和高分辨率的纹理。我们的论文贡献了面部地标和基于阴影的人体形状细化，语义纹理先验和新颖的纹理拼接策略，从而从单个视频中获得迄今为止最复杂的人类头像。大量的实验结果表明了该方法的鲁棒性和通用性。一项用户研究说明了它在身份保存、细节水平、现实主义和整体用户偏好方面优于最先进的技术。
We present a novel method for high detail-preserving human avatar creation from monocular video. A parameterized body model is refined and optimized to maximally resemble subjects from a video showing them from all sides. Our avatars feature a natural face, hairstyle, clothes with garment wrinkles, and high-resolution texture. Our paper contributes facial landmark and shading-based human body shape refinement, a semantic texture prior, and a novel texture stitching strategy, resulting in the most sophisticated-looking human avatars obtained from a single video to date. Numerous results show the robustness and versatility of our method. A user study illustrates its superiority over the state-of-the-art in terms of identity preservation, level of detail, realism, and overall user preference.