Face_recognition: Boost.Python.ArgumentError

[Jsmilemsj]

• Python version: 2.7.12
• Operating System: Ubuntu 16.04

Description

when I run "python facerec_from_webcam.py"，the following error occurred:

Traceback (most recent call last):
File "facerec_from_webcam.py", line 1, in
import face_recognition
File "/usr/local/lib/python2.7/dist-packages/face_recognition/__init__.py", line 7, in
from .api import load_image_file, face_locations, batch_face_locations, face_landmarks, face_encodings, compare_faces, face_distance
File "/usr/local/lib/python2.7/dist-packages/face_recognition/api.py", line 18, in
pose_predictor = dlib.shape_predictor(predictor_model)
Boost.Python.ArgumentError: Python argument types in
shape_predictor.__init__(shape_predictor, str)
did not match C++ signature:
__init__(boost::python::api::object, std::string)
__init__(_object*)

[ageitgey]

My guess is that something is wrong with the face_recognition_models files. Can you try re-installing the face_recognition_models package with pip and make sure no errors occur?

[Jsmilemsj]

OK, Thank you very much, I will try your suggestion.

At 2017-09-12 01:26:32, "Adam Geitgey" [e-mail] wrote:

My guess is that something is wrong with the face_recognition_models files. Can you try re-installing the face_recognition_models package with pip and make sure no errors occur?

—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub, or mute the thread.

[kenwoodjw]

I got same problem.some one can help?

File "/home/useease/kenwood/AlgorithmPlatform/venv/lib/python3.5/site-packages/face_recognition/api.py", line 18, in <module>
    pose_predictor = dlib.shape_predictor(predictor_model)
Boost.Python.ArgumentError: Python argument types in
    shape_predictor.__init__(shape_predictor, str)
did not match C++ signature:
    __init__(boost::python::api::object, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >)
    __init__(_object*)

[serhan-gul]

I also have the same problem with Python 2.7 and Boost 1.65. I'm using Anaconda and installed Boost through the channel conda-forge. When I try to import the face_recognition module, I get this error message:

>>> import face_recognition
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/home/guel/anaconda2/envs/py27/lib/python2.7/site-packages/face_recognition/__init__.py", line 7, in <module>
    from .api import load_image_file, face_locations, batch_face_locations, face_landmarks, face_encodings, compare_faces, face_distance
  File "/home/guel/anaconda2/envs/py27/lib/python2.7/site-packages/face_recognition/api.py", line 18, in <module>
    pose_predictor = dlib.shape_predictor(predictor_model)
Boost.Python.ArgumentError: Python argument types in
    shape_predictor.__init__(shape_predictor, str)
did not match C++ signature:
    __init__(boost::python::api::object, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >)
    __init__(_object*)

I also get the same error when I try to use dlib.cnn_face_detection_model_v1 so I think it is related to some incompatibility related to Boost.. Can someone help?

[jeremy-rutman]

same problem here , everything seems to have installed fine and the non-cnn face detector works fine

In [3]: import face_recognition
In [4]: import face_recognition_models
In [5]: face_recognition_models.cnn_face_detector_model_location()
Out[5]: '/usr/lib/python2.7/dist-packages/face_recognition_models/models/mmod_human_face_detector.dat'
In [6]: ls -l /usr/lib/python2.7/dist-packages/face_recognition_models/models/mmod_human_face_detector.dat
-rw-rw-r-- 1 jeremy jeremy 729940 Oct 30 14:45 /usr/lib/python2.7/dist-packages/face_recognition_models/models/mmod_human_face_detector.dat

[zengbing1992]

The same as me .Do you solve this problem?

prl@prl-System-Product-Name:~/Downloads/face_recognition-master/examples$ python find_faces_in_picture.py
Traceback (most recent call last):
File "find_faces_in_picture.py", line 2, in
import face_recognition
File "/usr/local/lib/python2.7/dist-packages/face_recognition/__init__.py", line 7, in
from .api import load_image_file, face_locations, batch_face_locations, face_landmarks, face_encodings, compare_faces, face_distance
File "/usr/local/lib/python2.7/dist-packages/face_recognition/api.py", line 18, in
pose_predictor = dlib.shape_predictor(predictor_model)
Boost.Python.ArgumentError: Python argument types in
shape_predictor.__init__(shape_predictor, str)
did not match C++ signature:
__init__(boost::python::api::object, std::string)
__init__(_object*)

[Wisgon]

I had met the same problem, and I solved it!!!

python:3.6(anaconda)
system:ubuntu17.10
dllib:19.7
use GPU: yes

Here is what I had done after the problem rose:
I reinstalled cmake following this guy:https://github.com/ageitgey/face_recognition/issues/120
Then, I rebuilded dlib-19.7 and recompiled it, then I reinstalled dlib, the command like this:

cd dlib
mkdir build
cd build
cmake .. -DDLIB_USE_CUDA=1 -DUSE_AVX_INSTRUCTIONS=0
cmake --build .
cd ..
python3 setup.py install  --yes DLIB_USE_CUDA  --no USE_AVX_INSTRUCTIONS 

If you don't use GPU, command like this:

cd dlib
mkdir build
cd build
cmake ..
cmake --build .
cd ..
python3 setup.py install 

then I reinstalled face_recognition:

pip uninstall face_recognition
pip install face_recognition

Finally, the problem solved,though I don't know why........

[idchangyifan]

Wisgon‘s way helps me solve the problem.
顶楼上

[Jsmilemsj]

Well done, thk u very much~

At 2018-01-02 15:01:53, "Wisgon Chen" [e-mail] wrote:

I had met the same problem, and I solved it!!!

python:3.6(anaconda)
system:ubuntu17.10
dllib:19.7
use GPU: yes

Here is what I had done after the problem rose:
I reinstalled cmake following this guy:#120
Then, I rebuilded dlib-19.7 and recompiled it, then I reinstalled dlib, the command like this:

cd dlib
mkdir build
cd build
cmake .. -DDLIB_USE_CUDA=1 -DUSE_AVX_INSTRUCTIONS=0
cmake --build .
cd ..
python3 setup.py install --yes DLIB_USE_CUDA --no USE_AVX_INSTRUCTIONS

If you don't use GPU, command like this:

cd dlib
mkdir build
cd build
cmake ..
cmake --build .
cd ..
python3 setup.py install

then I reinstalled face_recognition:

pip uninstall face_recognition
pip install face_recognition

Finally, the problem solved!

—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub, or mute the thread.

[tdjzyjy]

Wisgon solve my problem. If your boost is no problem, you can try it according to the method provided by Wisgon. Note that sudo pip3 install dlib==19.5.0 does not solve my problem, I finally reinstall dlib according to Wisgon