OpenCV实现人体姿态估计(人体关键点检测）OpenPose（图像、视频或调用摄像头均能实现）

图像、视频或调用摄像头均能实现OpenCV实现人体姿态估计(人体关键点检测）OpenPose人体姿态识别项目是美国卡耐基梅隆大学（CMU）基于卷积神经网络和监督学习并以Caffe为框架开发的开源库。可以实现人体动作、面部表情、手指运动等姿态估计。适用于单人和多人，具有极好的鲁棒性。是世界上首个基于深度学习的实时多人二维姿态估计应用。其理论基础来自Realtime Multi-Person 2D P

文章共5,792字 · 阅读需要大约20分钟

一键AI生成摘要，助你高效阅读

问答

北屿白

12147人浏览 · 2020-08-16 22:08:53

北屿白 · 2020-08-16 22:08:53 发布

图像、视频或调用摄像头均能实现OpenCV实现人体姿态估计(人体关键点检测）

OpenPose人体姿态识别项目是美国卡耐基梅隆大学（CMU）基于卷积神经网络和监督学习并以Caffe为框架开发的开源库。可以实现人体动作、面部表情、手指运动等姿态估计。适用于单人和多人，具有极好的鲁棒性。是世界上首个基于深度学习的实时多人二维姿态估计应用。
其理论基础来自Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields ，是CVPR 2017的一篇论文，作者是来自CMU感知计算实验室的曹哲（http://people.eecs.berkeley.edu/~zhecao/#top）,Tomas Simon,Shih-En Wei,Yaser Sheikh 。
人体姿态估计技术在体育健身、动作采集、3D试衣、舆情监测等领域具有广阔的应用前景，人们更加熟悉的应用就是抖音尬舞机。
在这里插入图片描述

OpenPose项目Github链接：https://github.com/CMU-Perceptual-Computing-Lab/openpose
OpenCV实现的Demo链接：[添加链接描述](https://github.com/PanJinquan/opencv-learning-tutorials/blob/master/opencv_dnn_pro/openpose-opencv/openpose_for_image_test.py)

1、实现原理

输入一幅图像，经过卷积网络提取特征，得到一组特征图，然后分成两个岔路，分别使用 CNN网络提取Part Confidence Maps 和 Part Affinity Fields；
在这里插入图片描述
得到这两个信息后，我们使用图论中的 Bipartite Matching（偶匹配）求出Part Association，将同一个人的关节点连接起来，由于PAF自身的矢量性，使得生成的偶匹配很正确，最终合并为一个人的整体骨架；
最后基于PAFs求Multi-Person Parsing—>把Multi-person parsing问题转换成graphs问题—>Hungarian Algorithm(匈牙利算法)
（匈牙利算法是部图匹配最常见的算法，该算法的核心就是寻找增广路径，它是一种用增广路径求二分图最大匹配的算法。)

2、实现神经网络

在这里插入图片描述

阶段一：VGGNet的前10层用于为输入图像创建特征映射。

阶段二：使用2分支多阶段CNN，其中第一分支预测身体部位位置（例如肘部，膝部等）的一组2D置信度图（S）。 如下图所示，给出关键点的置信度图和亲和力图 - 左肩。

在这里插入图片描述
第二分支预测一组部分亲和度的2D矢量场（L），其编码部分之间的关联度。如下图所示，显示颈部和左肩之间的部分亲和力。

阶段三： 通过贪心推理解析置信度和亲和力图，对图像中的所有人生成2D关键点。

3.OpenCV-OpenPose实现推理代码(图像)

# -*-coding: utf-8 -*-
"""
    @Project: python-learning-notes
    @File   : openpose_for_image_test.py
    @Author : panjq
    @E-mail : pan_jinquan@163.com
    @Date   : 2019-07-29 21:50:17
"""
 
import cv2 as cv
import os
import glob
 
BODY_PARTS = {"Nose": 0, "Neck": 1, "RShoulder": 2, "RElbow": 3, "RWrist": 4,
              "LShoulder": 5, "LElbow": 6, "LWrist": 7, "RHip": 8, "RKnee": 9,
              "RAnkle": 10, "LHip": 11, "LKnee": 12, "LAnkle": 13, "REye": 14,
              "LEye": 15, "REar": 16, "LEar": 17, "Background": 18}
 
POSE_PAIRS = [["Neck", "RShoulder"], ["Neck", "LShoulder"], ["RShoulder", "RElbow"],
              ["RElbow", "RWrist"], ["LShoulder", "LElbow"], ["LElbow", "LWrist"],
              ["Neck", "RHip"], ["RHip", "RKnee"], ["RKnee", "RAnkle"], ["Neck", "LHip"],
              ["LHip", "LKnee"], ["LKnee", "LAnkle"], ["Neck", "Nose"], ["Nose", "REye"],
              ["REye", "REar"], ["Nose", "LEye"], ["LEye", "LEar"]]
 
 
def detect_key_point(model_path, image_path, out_dir, inWidth=368, inHeight=368, threshhold=0.2):
    net = cv.dnn.readNetFromTensorflow(model_path)
    frame = cv.imread(image_path)
    frameWidth = frame.shape[1]
    frameHeight = frame.shape[0]
    scalefactor = 2.0
    net.setInput(
        cv.dnn.blobFromImage(frame, scalefactor, (inWidth, inHeight), (127.5, 127.5, 127.5), swapRB=True, crop=False))
    out = net.forward()
    out = out[:, :19, :, :]  # MobileNet output [1, 57, -1, -1], we only need the first 19 elements
    assert (len(BODY_PARTS) == out.shape[1])
    points = []
    for i in range(len(BODY_PARTS)):
        # Slice heatmap of corresponging body's part.
        heatMap = out[0, i, :, :]
        # Originally, we try to find all the local maximums. To simplify a sample
        # we just find a global one. However only a single pose at the same time
        # could be detected this way.
        _, conf, _, point = cv.minMaxLoc(heatMap)
        x = (frameWidth * point[0]) / out.shape[3]
        y = (frameHeight * point[1]) / out.shape[2]
        # Add a point if it's confidence is higher than threshold.
        points.append((int(x), int(y)) if conf > threshhold else None)
    for pair in POSE_PAIRS:
        partFrom = pair[0]
        partTo = pair[1]
        assert (partFrom in BODY_PARTS)
        assert (partTo in BODY_PARTS)
 
        idFrom = BODY_PARTS[partFrom]
        idTo = BODY_PARTS[partTo]
 
        if points[idFrom] and points[idTo]:
            cv.line(frame, points[idFrom], points[idTo], (0, 255, 0), 3)
            cv.ellipse(frame, points[idFrom], (3, 3), 0, 0, 360, (0, 0, 255), cv.FILLED)
            cv.ellipse(frame, points[idTo], (3, 3), 0, 0, 360, (0, 0, 255), cv.FILLED)
 
    t, _ = net.getPerfProfile()
    freq = cv.getTickFrequency() / 1000
    cv.putText(frame, '%.2fms' % (t / freq), (10, 20), cv.FONT_HERSHEY_SIMPLEX, 0.5, (0, 0, 0))
 
    cv.imwrite(os.path.join(out_dir, os.path.basename(image_path)), frame)
    cv.imshow('OpenPose using OpenCV', frame)
    cv.waitKey(0)
 
 
def detect_image_list_key_point(image_dir, out_dir, inWidth=480, inHeight=480, threshhold=0.3):
    image_list = glob.glob(image_dir)
    for image_path in image_list:
        detect_key_point(image_path, out_dir, inWidth, inHeight, threshhold)
 
 
if __name__ == "__main__":
    model_path = "pb/graph_opt.pb"
    # image_path = "body/*.jpg"
    out_dir = "result"
    # detect_image_list_key_point(image_path,out_dir)
    image_path = "./test.jpg"
    detect_key_point(model_path, image_path, out_dir, inWidth=368, inHeight=368, threshhold=0.05)

在这里插入图片描述

4.OpenCV-OpenPose实现推理代码（视频或摄像头）

import cv2
import numpy as np
import argparse

parser = argparse.ArgumentParser()
parser.add_argument('--input', help='Path to image or video. Skip to capture frames from camera')
parser.add_argument('--thr', default=0.2, type=float, help='Threshold value for pose parts heat map')
parser.add_argument('--width', default=270, type=int, help='Resize input to specific width.')
parser.add_argument('--height', default=480, type=int, help='Resize input to specific height.')

args = parser.parse_args()

BODY_PARTS = {"Nose": 0, "Neck": 1, "RShoulder": 2, "RElbow": 3, "RWrist": 4,
              "LShoulder": 5, "LElbow": 6, "LWrist": 7, "RHip": 8, "RKnee": 9,
              "RAnkle": 10, "LHip": 11, "LKnee": 12, "LAnkle": 13, "REye": 14,
              "LEye": 15, "REar": 16, "LEar": 17, "Background": 18}

POSE_PAIRS = [["Neck", "RShoulder"], ["Neck", "LShoulder"], ["RShoulder", "RElbow"],
              ["RElbow", "RWrist"], ["LShoulder", "LElbow"], ["LElbow", "LWrist"],
              ["Neck", "RHip"], ["RHip", "RKnee"], ["RKnee", "RAnkle"], ["Neck", "LHip"],
              ["LHip", "LKnee"], ["LKnee", "LAnkle"], ["Neck", "Nose"], ["Nose", "REye"],
              ["REye", "REar"], ["Nose", "LEye"], ["LEye", "LEar"]]

inWidth = args.width
inHeight = args.height


net = cv2.dnn.readNetFromTensorflow("graph_opt.pb")

#cap = cv2.VideoCapture(args.input if args.input else 0)
#cap = cv2.VideoCapture('action.mp4')
cap = cv2.VideoCapture(0)

fourcc = cv2.VideoWriter_fourcc(*'XVID')
output = cv2.VideoWriter('output.avi', fourcc, 5.0, (int(cap.get(3)),int(cap.get(4))))
'''注意：在这部分若要调用摄像头进行实时图像检测并保存视频到本地，需要将上面三行的参数进行更改，并保证最终的保存视频的尺寸大小与输出图像大小相同。'''
while True:
    hasFrame, frame = cap.read()

    frameWidth = frame.shape[1]
    frameHeight = frame.shape[0]

    net.setInput(cv2.dnn.blobFromImage(frame, 1.0, (inWidth, inHeight), (127.5, 127.5, 127.5), swapRB=True, crop=False))
    out = net.forward()
    out = out[:, :19, :, :]  # MobileNet output [1, 57, -1, -1], we only need the first 19 elements

    assert (len(BODY_PARTS) == out.shape[1])

    points = []
    for i in range(len(BODY_PARTS)):
        # Slice heatmap of corresponging body's part.
        heatMap = out[0, i, :, :]

        # Originally, we try to find all the local maximums. To simplify a sample
        # we just find a global one. However only a single pose at the same time
        # could be detected this way.
        _, conf, _, point = cv2.minMaxLoc(heatMap)
        x = (frameWidth * point[0]) / out.shape[3]
        y = (frameHeight * point[1]) / out.shape[2]
        # Add a point if it's confidence is higher than threshold.
        points.append((int(x), int(y)) if conf > args.thr else None)

    for pair in POSE_PAIRS:
        partFrom = pair[0]
        partTo = pair[1]
        assert (partFrom in BODY_PARTS)
        assert (partTo in BODY_PARTS)

        idFrom = BODY_PARTS[partFrom]
        idTo = BODY_PARTS[partTo]

        if points[idFrom] and points[idTo]:
            cv2.line(frame, points[idFrom], points[idTo], (0, 255, 0), 3)
            cv2.ellipse(frame, points[idFrom], (3, 3), 0, 0, 360, (0, 0, 255), cv2.FILLED)
            cv2.ellipse(frame, points[idTo], (3, 3), 0, 0, 360, (0, 0, 255), cv2.FILLED)

    t, _ = net.getPerfProfile()
    freq = cv2.getTickFrequency() / 1000
    cv2.putText(frame, '%.2fms' % (t / freq), (10, 20), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 0, 0))

    cv2.imshow('OpenPose using OpenCV', frame)
    output.write(frame)

    if cv2.waitKey(1) & 0xFF == ord('q'):
        break

cap.release()
output.release()
cv2.destroyAllWindows()