# face_3d

**Repository Path**: numerical_aggregation_research/face_3d

## Basic Information

- **Project Name**: face_3d
- **Description**: 3d人脸分析
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2019-11-12
- **Last Updated**: 2021-11-03

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# 3d人脸项目

# 安装

## 1 编译安装（不推荐）
## 环境变量
```shell
D:\Program Files\PCL 1.9.1\bin
D:\Program Files\PCL 1.9.1\3rdParty\Boost\lib
D:\Program Files\PCL 1.9.1\3rdParty\FLANN\bin
D:\Program Files\PCL 1.9.1\3rdParty\VTK\bin
D:\Program Files\PCL 1.9.1\3rdParty\Qhull\bin
D:\Program Files\OpenNI2\Lib
D:\Program Files\OpenNI2\Redist
D:\Program Files\OpenNI2\Tools
```
## 2 直接安装（推荐）
```shell
1 下载一个 python-pcl 的 .whl 文件
https://ci.appveyor.com/project/Sirokujira/python-pcl-iju42/build/job/dqefbpxo8h9e0yyb/artifacts

2 安装whl

3 安装其他包
pip install numpy opencv-python cython scikit-image -i https://pypi.douban.com/simple

4 测试

import cv2


depth = cv2.imread("depth.png",-1) 

import pcl
cloud = pcl.PointCloud() # 存储图像的点云
rows = len(depth)
cols = len(depth[0])
pointcloud = []

#camera_factor = 1
#camera_cx = 0
#camera_cy = 0
camera_fx = 588.03
camera_fy = 587.07

for m in range(0, rows):
    for n in range(0, cols):
        d = depth[m][n][0] + depth[m][n][1]*256
        if d == 0:
            pass
        else:
            z = float(d)
            x = n * z / camera_fx
            y = m * z / camera_fy
            points = [x, y, z]
            pointcloud.append(points)
import numpy as np
pointcloud = np.array(pointcloud, dtype = np.float32)
cloud.from_array(pointcloud)
pcl.save(cloud, "cloud.pcd", format = 'pcd')
```

# 参考
## 仓库
- [pcl](https://github.com/PointCloudLibrary/pcl)
- [python-pcl](https://github.com/strawlab/python-pcl)
- [Face3D-Pytorch 人脸识别](https://github.com/XingwXiong/Face3D-Pytorch)
    - [3dmm](https://blog.csdn.net/likewind1993/article/details/81455882)
- [face3d 三维人脸处理](https://github.com/YadiraF/face3d)
- [pclpy](https://github.com/davidcaron/pclpy)

## 资料
- [Win10下python-pcl库的安装配置](https://blog.csdn.net/find_god/article/details/101106373)
- [windows 环境安装 python-pcl 【极简方案】](http://www.pianshen.com/article/7372313057/)
- [https://ci.appveyor.com/project/Sirokujira/python-pcl-iju42](https://ci.appveyor.com/project/Sirokujira/python-pcl-iju42)

## 应用
- [3D人脸结合“相貌认知学”，上海影子智能：让整形效果“精确预览”](http://www.sohu.com/a/299857190_474144)
- [上海影子智能](http://www.myreal3d.com/)