一、数据集概述
人类行为数据集由外国人工智能工程师收集。我衷心感谢开拓者们的努力。该数据集的难点在于人类行为的多样性和背景的多重干扰，以及较多的类内变异。
[En]
The human behavior data set is collected by foreign artificial intelligence engineers. I sincerely thank the pioneers for their efforts. The difficulty of this data set lies in the diversity of human behavior and the multi-interference of the background, as well as more intra-class variation.
数据集中的训练部分为15000张图片，测试集为3000张图片。
收集到的人类行为分为15类。
[En]
The collected human behaviors are divided into 15 categories.
分为打电话、鼓掌、骑行、跳舞、喝酒、吃饭、打架、拥抱、大笑、听音乐、跑步、坐着、睡觉、发短信、使用笔记本电脑等。
[En]
It is divided into phone call, applause, cycling, dancing, drinking, eating, fighting, hugging, laughing, listening to music, running, sitting, sleeping, texting, using laptop and so on.
数据集下载地址：
链接：https://pan.baidu.com/s/1C4zXiiC1zcxo1D7IcZqOFA
提取码：bor5
跑步样例图片
二、自定义模型参考代码
1、导入必要的包
import os
import glob
import numpy as np
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
from tensorflow.keras.layers import Dense
from tensorflow.keras.models import Model
from keras.preprocessing.image import ImageDataGenerato
Original: https://blog.csdn.net/bashendixie5/article/details/124501573
Author: 坐望云起
Title: 机器学习笔记 – 基于tensorflow的人类行为检测
原创文章受到原创版权保护。转载请注明出处：https://www.johngo689.com/509582/
转载文章受原作者版权保护。转载请注明原作者出处！