openface+conda python.10排坑记录_openface python

任务

发现openface好像还有一个python版本的，跑起来试试。

注意

我希望把博客写得既能帮自己从实践中总结经验，又能帮助做同样事情的同学快速找到参考。如果你有任何关于博客内容和排版的建议，请在评论区告诉我，谢谢~

成功流程

按照参考博客来。

1. 我的系统是20.04，新建了一个python3.10的conda环境；

conda create -n OPENFACE python=3.10
conda activate OPENFACE
1
2

2. 手动安装requirements.txt的python依赖，不要根据他的版本要求自动下载，大概率出错，我直接无视版本安装没问题。中间缺什么依赖安装什么依赖。
3. 根据setup.py安装openface
4. 打开python环境尝试import openface，把缺失的依赖安装好
5. 在openface目录下复制上述博客的代码，做少许改动，然后运行。

踩坑回顾

1. requirements依赖错误

症状描述

执行：

pip3 install -r requirements.txt
1

报错：

Collecting pandas<0.18,>=0.13 (from -r requirements.txt (line 3))
  Using cached pandas-0.17.1.zip (7.7 MB)
  Preparing metadata (setup.py) ... error
  error: subprocess-exited-with-error
  
  × python setup.py egg_info did not run successfully.
  │ exit code: 1
  ╰─> [15 lines of output]
      /tmp/pip-install-wc34lmil/pandas_25f6faa563c9402db75b67e3a03f6b26/setup.py:30: DeprecationWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html
        import pkg_resources
      /home/lcy-magic/anaconda3/envs/OPENFACE/lib/python3.10/site-packages/setuptools/__init__.py:80: _DeprecatedInstaller: setuptools.installer and fetch_build_eggs are deprecated.
      !!
      
              ********************************************************************************
              Requirements should be satisfied by a PEP 517 installer.
              If you are using pip, you can try `pip install --use-pep517`.
              ********************************************************************************
      
      !!
        dist.fetch_build_eggs(dist.setup_requires)
      error in pandas setup command: 'install_requires' must be a string or list of strings containing valid project/version requirement specifiers; Expected end or semicolon (after version specifier)
          pytz >= 2011k
               ~~~~~~~^
      [end of output]
  
  note: This error originates from a subprocess, and is likely not a problem with pip.
error: metadata-generation-failed

× Encountered error while generating package metadata.
╰─> See above for output.

note: This is an issue with the package mentioned above, not pip.
hint: See above for details.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33

解决办法

尝试升级pip

pip3 install --upgrade pip
1

还是同样报错。
因为报错建议使用-use-pep517，是一种不同于setuptools的新的包构建方式。所以估计还是包的依赖问题。先尝试升级setuptools：

pip3 install --upgrade setuptools
1

报错：

ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
rosdep 0.22.2 requires catkin-pkg>=0.4.0, which is not installed.
rosdep 0.22.2 requires PyYAML>=3.1, which is not installed.
rosdistro 0.9.0 requires catkin-pkg, which is not installed.
rosdistro 0.9.0 requires PyYAML, which is not installed.
rosinstall 0.7.8 requires catkin-pkg, which is not installed.
rosinstall 0.7.8 requires pyyaml, which is not installed.
rosinstall-generator 0.1.23 requires catkin-pkg>=0.1.28, which is not installed.
rosinstall-generator 0.1.23 requires PyYAML, which is not installed.
Successfully installed setuptools-69.0.3
1
2
3
4
5
6
7
8
9
10

问题不大，成功安装了。只是缺少点依赖。全安装上就行了：

pip3 install catkin-pkg PyYAML rospkg
1

但还是同样报错。
那试试用报错建议的方法吧：

pip3 install --use-pep517 -r requirements.txt
1

结果还是不行，报错太长就不贴出来了，应该还是依赖问题。
按照参考博客和里面评论的建议做了，也还是不行。
然后按照参考博客指定setuptools版本，虽然pandas解决了，但后面的包还是有问题。但我当时用的是python3 -m pip install --upgrade pip，但我用python3其实对应的是系统的解释器，我的conda的解释器要用python，改成python -m pip install --upgrade pip说不定可以。反正我已经解决了，就不尝试了。

最后一气之下，决定不按照requirements的版本来了，直接都上最新版本：

pip3 install numpy scipy pandas scikit-learn nose nolearn
1

这样暂时都没问题，看看后面会不会出依赖问题吧。

2. 依赖缺失错误

症状描述

用setup.py安装openface：

python setup.py install
1

然后进入python，去import openface，会提示没有cv2和dlib。

解决办法

这都是小问题。直接安装上就行了：

pip3 install opencv-python dlib
1

成功：

3. 读不了摄像头

症状描述

我在openface目录下，复制了参考博客的程序，并运行，报错：

[ WARN:0@0.541] global cap_v4l.cpp:997 open VIDEOIO(V4L2:/dev/video1): can't open camera by index
[ERROR:0@0.660] global obsensor_uvc_stream_channel.cpp:159 getStreamChannelGroup Camera index out of range
1
2

解决方法

也是很简单问题，就是读取摄像头的索引是1，我电脑就一个摄像头，编号就是0，改一下就行。然后再加一条语句，可以调整窗口大小。修改后的代码如下：

import threading
import cv2
import os
import openface
 
 
fileDir = os.path.dirname(os.path.realpath(__file__))
modelDir = os.path.join(fileDir, 'models')
dlibModelDir = os.path.join(modelDir, 'dlib')
 
align = openface.AlignDlib(os.path.join(dlibModelDir, "shape_predictor_68_face_landmarks.dat"))
 
 
class OpcvCapture(threading.Thread):
    def __init__(self, win_name, cam_name):
        super().__init__()
        self.cam_name = cam_name
        self.win_name = win_name
        # self.frame = np.zeros([400, 400, 3], np.uint8)
 
    def run(self):
        capture = cv2.VideoCapture(self.cam_name)
        while (True):
            # 获取一帧
            ret, frame = capture.read()
            # 获取的帧送入检测，绘制检测结果后返回,自拍模式做镜像
            show_img = self._detector(frame, mirror=True)
            cv2.namedWindow(self.win_name, cv2.WINDOW_NORMAL)
            cv2.imshow(self.win_name, show_img)
            cv2.waitKey(1)
 
    def _detector(self, frame, mirror=False):
        show_img = cv2.flip(frame, flipCode=1) if mirror else frame
        rects = align.getAllFaceBoundingBoxes(show_img)
        if len(rects) > 0:
            bb = align.findLandmarks(show_img, rects[0])
            for pt in bb:
                cv2.circle(show_img, pt, 3, [0, 0, 255], thickness=-1)
        return show_img
 
 
if __name__ == "__main__":
    camera1 = OpcvCapture("camera1", 0)
    camera1.start()
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44

感谢源代码来源的分享，非常感谢。

最后效果