关键点检测——热图标签（Ground Truth）构建

最新推荐文章于 2024-10-10 15:27:42 发布

zouxiaolv

最新推荐文章于 2024-10-10 15:27:42 发布

阅读量5.8k

点赞数 3

分类专栏：人体姿态估计文章标签：关键点检测——热图标签（Ground Truth）构建

本文链接：https://blog.csdn.net/zouxiaolv/article/details/100688761

版权

人体姿态估计专栏收录该内容

64 篇文章 13 订阅

订阅专栏

参考文献：

https://blog.csdn.net/zziahgf/article/details/79704788

https://zhuanlan.zhihu.com/p/37933909

https://blog.csdn.net/u013841196/article/details/85643310#comments

$\left ( x_{k}, y_{k} \right )$ 是真实标签的的坐标。控制该函数的左右移动情况

$\left ( x, y\right )$ 是热图的feature map 中每一个像素位置

$\sigma$ ：控制着该函数的胖瘦，也就是宽窄，涉及的范围多大，即径向范围

代码实现：

import time
import numpy as np
import cv2
import matplotlib.pyplot as plt


def CenterLabelHeatMap(img_width, img_height, c_x, c_y, sigma):
    X1 = np.linspace(1, img_width, img_width)
    Y1 = np.linspace(1, img_height, img_height)
    [X, Y] = np.meshgrid(X1, Y1)
    X = X - c_x
    Y = Y - c_y
    D2 = X * X + Y * Y
    E2 = 2.0 * sigma * sigma
    Exponent = D2 / E2
    heatmap = np.exp(-Exponent)
    return heatmap

#################另一种生成方法############


# Compute gaussian kernel
def CenterGaussianHeatMap(img_height, img_width, c_x, c_y, variance):
    gaussian_map = np.zeros((img_height, img_width))
    for x_p in range(img_width):
        for y_p in range(img_height):
            dist_sq = (x_p - c_x) * (x_p - c_x) + \
                      (y_p - c_y) * (y_p - c_y)
            exponent = dist_sq / 2.0 / variance / variance
            gaussian_map[y_p, x_p] = np.exp(-exponent)
    return gaussian_map


image_file = 'test.jpg'
img = cv2.imread(image_file)
img = img[:,:,::-1]

height, width,_ = np.shape(img)
cy, cx = height/2.0, width/2.0

start = time.time()
heatmap1 = CenterLabelHeatMap(width, height, cx, cy, 21)
t1 = time.time() - start

start = time.time()
heatmap2 = CenterGaussianHeatMap(height, width, cx, cy, 21)
t2 = time.time() - start

print(t1, t2)

plt.subplot(1,2,1)
plt.imshow(heatmap1)
plt.subplot(1,2,2)
plt.imshow(heatmap2)
plt.show()

print('End.')