Posture detection by openpose from USB camera image using python code

Purpose

It is a memorandum when posture detection by openpose is performed from a USB camera image using python code.

Preparation

Ubuntu16.04 PC Webcam

Refer to the article Openpose PYTHON API and proceed with the preparation until the test code (1_extract_pose.py) is moved.

The version of CUDA / cuDNN uses the following. 　CUDA ver8.0, 　cuDNN ver6.0 I tried running OpenPose. was used as a reference.

code

test.py

# From Python
# It requires OpenCV installed for Python
import sys
import cv2
import os
from sys import platform
import cv2

# Remember to add your installation path here
# Option a
dir_path = os.path.dirname(os.path.realpath(__file__))
if platform == "win32": sys.path.append(dir_path + '/../../python/openpose/');
else: sys.path.append('../../python');
# Option b
# If you run `make install` (default path is `/usr/local/python` for Ubuntu), you can also access the OpenPose/python module from there. This will install OpenPose and the python library at your desired installation path. Ensure that this is in your python path in order to use it.
# sys.path.append('/usr/local/python')

# Parameters for OpenPose. Take a look at C++ OpenPose example for meaning of components. Ensure all below are filled
try:
    from openpose import *
except:
    raise Exception('Error: OpenPose library could not be found. Did you enable `BUILD_PYTHON` in CMake and have this Python script in the right folder?')
params = dict()
params["logging_level"] = 3
params["output_resolution"] = "-1x-1"
params["net_resolution"] = "-1x368"
params["model_pose"] = "BODY_25"
params["alpha_pose"] = 0.6
params["scale_gap"] = 0.3
params["scale_number"] = 1
params["render_threshold"] = 0.05
# If GPU version is built, and multiple GPUs are available, set the ID here
params["num_gpu_start"] = 0
params["disable_blending"] = False
# Ensure you point to the correct path where models are located
#params["default_model_folder"] = dir_path + "/../../../models/"
params["default_model_folder"] = "../../../models/"
# Construct OpenPose object allocates GPU memory
openpose = OpenPose(params)

cap = cv2.VideoCapture(0)

while 1:
    # Read new image

    ret, img = cap.read()
    # img = cv2.imread("../../../examples/media/COCO_val2014_000000000192.jpg ")

    # Output keypoints and the image with the human skeleton blended on it
    keypoints, output_image = openpose.forward(img, True)

    # Print the human pose keypoints, i.e., a [#people x #keypoints x 3]-dimensional numpy object with the keypoints of all the people on that image
    print(keypoints)
    # Display the image
    cv2.imshow("output", output_image)
    cv2.waitKey(15)

cap.release()
cv2.destroyAllWindows()

test

It is OK if the result of posture detection by openpose is output from the image input from the WEB camera below.

/openpose/build/examples/tutorial_python$ python test.py 

Coding Error measures

Prototxt file not found: /usr/local/python/openpose/../../../models error

$ python 1_extract_pose.py
Error:
Prototxt file not found: /usr/local/python/openpose/../../../models/pose/body_25/pose_deploy.prototxt.
Possible causes:
	1. Not downloading the OpenPose trained models.
	2. Not running OpenPose from the same directory where the `model` folder is located.
	3. Using paths with spaces.

Because models cannot be referenced from the runtime directory It's OK if you modify 1_extract_pose.py so that openpose / models can refer to it.

#params["default_model_folder"] = dir_path + "/../../../models/"
params["default_model_folder"] = "../../../models/"

reference

Openpose PYTHON API