Dress-Me-Up
收藏arXiv2024-01-06 更新2024-08-06 收录
下载链接：
http://arxiv.org/abs/2401.03108v1
下载链接
链接失效反馈官方服务：
资源简介：
Dress-Me-Up数据集是首个用于非参数化真实3D服装调整的基准和数据集。该数据集包含255件真实世界的服装，这些服装由15个不同的主体在5种不同的姿势下穿着，使用多视角RGBD捕捉系统捕获。数据集旨在评估3D非参数化服装调整框架，展示了对非参数化服装和人体化身在现有最先进方法上的优越调整质量，作为非参数化3D服装调整的第一个基准。
The Dress-Me-Up dataset is the first benchmark and dataset dedicated to non-parametric real-world 3D clothing adjustment. It contains 255 real-world garments worn by 15 distinct human subjects across five different poses, captured using a multi-view RGBD capture system. This dataset is designed to evaluate 3D non-parametric clothing adjustment frameworks, and demonstrates superior adjustment quality for non-parametric clothing and human avatars compared to existing state-of-the-art methods, serving as the first benchmark for non-parametric 3D clothing adjustment.
提供机构：
国际信息技术研究所
创建时间：
2024-01-06
