3D human skeleton 数据集简易可视化

从一个博客参考
我使用的数据集只有骨骼数据（三维坐标）

def draw3Dpose(pose_3d, ax, lcolor="#3498db", rcolor="#e74c3c", add_labels=False):  # blue, orange# 骨架连接顺序：起始关节，终止关节，左右关节标识(1 left 0 right)IRDS_connect_dict = [[0, 16, 0], [16, 17, 0], [17, 18, 0], [18, 19, 0], [0, 12, 1], [12, 13, 1], [13, 14, 1],[14, 15, 1], [0, 1, 0], [1, 20, 0], [20, 2, 1], [2, 3, 1], [20, 8, 0], [8, 9, 0], [9, 10, 0],[10, 11, 0],[11, 24, 0], [11, 23, 0], [20, 4, 1], [4, 5, 1], [5, 6, 1], [6, 7, 1], [7, 22, 1], [7, 21, 1]]# for i in human36m_connectivity_dict:for i in IRDS_connect_dict:x, z, y = [np.array([pose_3d[i[0], j], pose_3d[i[1], j]]) for j in range(3)]ax.plot(x, y, z, lw=2, c=lcolor if i[2] else rcolor)RADIUS = 0.750  # space around the subjectxroot, yroot, zroot = pose_3d[5, 0], pose_3d[5, 1], pose_3d[5, 2]ax.set_xlim3d([-RADIUS + xroot, RADIUS + xroot])ax.set_ylim3d([0, 2 * RADIUS + zroot])ax.set_zlim3d([-RADIUS + yroot, RADIUS + yroot])ax.set_xlabel("x")ax.set_ylabel("y")ax.set_zlabel("z")def view_skeleton(path):
"""处理数据，变成[frame,pointsnum,3]格式"""specific_3d_skeleton = np.loadtxt(path, dtype=np.float, encoding='utf-8', delimiter=',')new_specific_3d_skeleton = []for i in range(len(specific_3d_skeleton)):frame = []frame = np.reshape(specific_3d_skeleton[i], (25, 3))new_specific_3d_skeleton.append(frame)new_specific_3d_skeleton = np.array(new_specific_3d_skeleton)fig = plt.figure()ax = fig.add_subplot(111, projection='3d')plt.ion()#展示多图i = 0#死循环，一直展示动图while i < new_specific_3d_skeleton.shape[0]:ax.lines = []draw3Dpose(new_specific_3d_skeleton[i], ax)plt.pause(0.001)i += 1if i == new_specific_3d_skeleton.shape[0]:i = 0plt.ioff()plt.show()
if __name__ == '__main__':""" after feedback action replay """pat_action = r'E:\pythonFile\myrehab\standard\0\101_18_0_4_1_stand.txt'view_skeleton(pat_action)print("...end...")

骨架连接顺序以及xyz轴可根据自己数据进行调整
如我刚开始用别人的代码画出来的人物是躺着而不是站着，调整坐标轴即可

3D human skeleton 数据集简易可视化相关推荐

论文阅读 Semantic Graph Convolutional Networks for 3D Human Pose Regression
Semantic Graph Convolutional Networks for 3D Human Pose Regression 使用语义图卷积网络对三维人体姿态进行回归 Abstract 在 ...
行为识别论文阅读（2）——3D Human Sensing, Action and Emotion Recognition in Robot Assisted Therapy of Children
行为识别论文阅读(2)--3D Human Sensing, Action and Emotion Recognition in Robot Assisted Therapy of Children ...
2D/3D人体姿态估计（2D/3D Human Pose Estimation）
1. 基本概念算法改进入口网络设计特征流损失函数数据集的重要性:只要有一个好的.针对性的数据集,问题都可以解决过集成新一代AutoML技术,可降低算法试错成本人体姿态估计(Human P ...
HybrIK: A Hybrid Analytical-Neural Inverse Kinematics Solution for 3D Human Pose and Shape Estimatio
CVPR 2021 [paper] [code] 摘要 (1)基于模型的三维姿态和形状估计方法通过估计人体的几个参数来重建完整的三维网格.然而,抽象参数的学习是一个高度非线性的过程,且图像与模型存在偏 ...
3D鸟类重建—数据集、模型以及从单视图恢复形状
点击上方"小白学视觉",选择"星标" 干货第一时间送达代码.项目.论文地址:在公众号「计算机视觉工坊」,后台回复「3D鸟类重建」,即可直接下载. 概述动物姿 ...
3D Human相关研究总结：人体、姿态估计、人体重建等
©PaperWeekly 原创 · 作者|张莹单位|腾讯本文简要介绍与 3D 数字人相关的研究,包括常用 3D 表示.常用 3D 人体模型.3D 人体姿态估计,带衣服 3D 人体重建,3D 衣服建 ...
论文笔记--3D human pose estimation in video with temporal convolutions and semi-supervised training
3D human pose estimation in video with temporal convolutions and semi-supervised training(利用时间卷积和半监督 ...
论文阅读 [TPAMI-2022] Locally Connected Network for Monocular 3D Human Pose Estimation
论文阅读 [TPAMI-2022] Locally Connected Network for Monocular 3D Human Pose Estimation 论文搜索(studyai.com) ...
视频中的3D人体姿态估计(3D human pose estimation in vide)--------Facebook research: VideoPose3D
视频中的3D人体姿态估计(3D human pose estimation in video) 解读Facebook AI Research in CVPR2019: <3D human pos ...

3D human skeleton 数据集简易可视化

3D human skeleton 数据集简易可视化相关推荐

最新文章

热门文章