This repository provides the EBStore dataset of EMS.
EMS: EMS: 3D Eyebrow Modeling from Single-view Images
Chenghong Li*, Leyang Jin*, Yujian Zheng, Yizhou Yu# and Xiaoguang Han#
EBStore is the first high-quality 3D synthetic eyebrow dataset containing 400 high-quality 3D models. These eyebrow models are manually created by artists using particle edit mode in Blender, guided by both the high-resolution texture map of the eyebrow region and the front-facing image of the individual.
To download EBStore data, please visit OneDrive. After unzipping the downloaded file, you will find the following directory structure:
├── EBStore_blender_data
│ ├── 1
│ │ ├── 1_uv.jpg
│ │ ├── 1.blend
│ │ ├── 1.obj
│ │ ├── h.mtl
│ │ ├── h.obj
│ │ ├── m.mtl
│ │ └── m.obj
│ ├── 2
│ │ ├── 2_uv.jpg
│ │ ├── 2.blend
│ │ ├── 2.obj
│ │ ├── h.mtl
│ │ ├── h.obj
│ │ ├── m.mtl
│ │ └── m.obj
├── ...
We test our code on blender-3.4.0 and blender-3.6.14.
python hair2npy.py --input ./EBStore_blender_data --output ./strand_npy_rekey
python npy2obj.py --input ./strand_npy_rekey --output ./strand_obj_rekey
The data and the code are released under the Attribution-NonCommercial 4.0 International License.
Copyright (c) 2024
If you find the data and code of this work helpful to your research, please consider citing:
@article{li2023ems,
title={EMS: 3D Eyebrow Modeling from Single-view Images},
author={Li, Chenghong and Jin, Leyang and Zheng, Yujian and Yu, Yizhou and Han, Xiaoguang},
journal={ACM Transactions on Graphics (TOG)},
volume={42},
number={6},
pages={1--19},
year={2023},
publisher={ACM New York, NY, USA}
}