按照kinetics-skeleton规范自制st-gcn数据集

此处参考了博客：https://blog.csdn.net/Lujiahao98689/article/details/121447175
根据自己电脑配置
[win10 python3.7 pytorch1.10.1 cuda11.3]
进行细微修改，日后备用，特此记录

第一步：裁剪视频，将视频裁剪到8到10秒钟
此处可以找辅助的软件进行裁剪

第二步：视频左右翻转，扩充数据集
代码如下：

## flip_horizontal_video.py
import os
# 知识点1，要避免出现‘Cannot find installation of real FFmpeg (which comes with ffprobe)‘
# 在import skvideo.io之前，先skvideo.setFFmpegPath('ffmpeg所在的文件位置')
import skvideo
skvideo.setFFmpegPath('D:/INSTALLER/ffmpeg-5.0.1-essentials_build/bin')
import skvideo.io
import cv2

if __name__ == '__main__':
    originvideo_file = 'D:/VIDEOS/running on treadmill/'
    videos_file_names = os.listdir(originvideo_file)
    # 1. 左右镜像翻转视频
    for file_name in videos_file_names:
        video_path = '{}{}'.format(originvideo_file, file_name)

        name_without_suffix = file_name.split('.')[0]
        outvideo_path = '{}{}_mirror.mp4'.format(originvideo_file, name_without_suffix)

        writer = skvideo.io.FFmpegWriter(outvideo_path)
                                         # outputdict={'-f': 'mp4', '-vcodec': 'libx264', '-r': '30'})
        reader = skvideo.io.FFmpegReader(video_path)
        for frame in reader.nextFrame():
            frame_mirror = cv2.flip(frame, 1)
            writer.writeFrame(frame_mirror)
        writer.close()
        print('{} mirror success'.format(file_name))

    print('the video in are all mirrored')
    print('-------------------------------------------------------')

第三步：先将视频统一调整成340x256，调用cmd指令，使用openpose中的OpenPoseDemo.exe给每个视频生成对应的json文件

## 
#!/usr/bin/env python
import os
import argparse
import json
import shutil

import numpy as np
import torch
import skvideo
skvideo.setFFmpegPath('D:/INSTALLER/ffmpeg-5.0.1-essentials_build/bin')
import skvideo.io

import tools
import tools.utils as utils


label_name, label_no = 'running', 0

class PreProcess():
    """
        利用openpose提取自建数据集的骨骼点数据
    """

    def start(self):

        work_dir = './st-gcn-master'

        # gongfu_filename = gongfu_filename_list[process_index]
        # # 标签信息
        # labelgongfu_name = 'xxx_{}'.format(process_index)
        # label_no = process_index

        #################################################################
        # 此处修改成要制作的video文件
        #################################################################

        # 视频所在文件夹 D:\VIDEOS\test_flip
        originvideo_file = 'D:/VIDEOS/running on treadmill/'
        # resized视频输出文件夹
        name_resizecideo_file = 'resize_running_on_treadmill'
        resizedvideo_file = 'D:/VIDEOS/{}/'.format(name_resizecideo_file)
        os.makedirs(resizedvideo_file, exist_ok=True)

        videos_file_names = os.listdir(originvideo_file)

        # 1. Resize文件夹下的视频到340x256 30fps
        for file_name in videos_file_names:
            video_path = '{}{}'.format(originvideo_file, file_name)
            outvideo_path = '{}{}'.format(resizedvideo_file, file_name)
            writer = skvideo.io.FFmpegWriter(outvideo_path, outputdict={'-s': '340x256'})
                                             # outputdict={'-f': 'mp4', '-vcodec': 'libx264', '-s': '340x256',
                                             #             '-r': '30'})
            reader = skvideo.io.FFmpegReader(video_path)
            for frame in reader.nextFrame():
                writer.writeFrame(frame)
            writer.close()
            print('{} resize success'.format(file_name))

        # 2. 利用openpose提取每段视频骨骼点数据
        resizedvideos_file_names = os.listdir(resizedvideo_file)
        for file_name in resizedvideos_file_names:
            if not file_name.endswith('mp4'):
                continue
            outvideo_path = '{}{}'.format(resizedvideo_file, file_name)

            # openpose = '{}/examples/openpose/openpose.bin'.format(self.arg.openpose)
            # 'D:/Pycharm_project/Pose/openpose-master/build/bin'
            openpose_path = f'D:/Pycharm_project/Pose/openpose_all/openpose'
            openpose = 'bin\\OpenPoseDemo.exe'
            video_name = os.path.splitext(file_name)[0]
            output_snippets_dir = 'D:\\VIDEOS\\{}\\snippets\\{}'.format(name_resizecideo_file, video_name)
            output_sequence_dir = 'D:\\VIDEOS\\{}\\data'.format(name_resizecideo_file)
            os.makedirs(output_snippets_dir, exist_ok=True)
            os.makedirs(output_sequence_dir, exist_ok=True)
            output_sequence_path = '{}/{}.json'.format(output_sequence_dir, video_name)

            # # 这一段莫名其妙啊
            # label_name_path = '{}/resource/kinetics_skeleton/label_name_gongfu.txt'.format(work_dir)
            # with open(label_name_path) as f:
            #     label_name = f.readlines()
            #     label_name = [line.rstrip() for line in label_name]

            # pose estimation
            openpose_args = dict(
                video=outvideo_path,    # 'D:/VIDEOS/resize_test_flip/data'
                write_json=output_snippets_dir,
                display=0,
                render_pose=0,
                model_pose='COCO')
            command_line = 'd:&cd ' + openpose_path + '& '
            command_line += openpose +' '+ ' '.join(['--{} {}'.format(k, v) for k, v in openpose_args.items()])
            # D:/Pycharm_project/Pose/openpose_all/openpose/bin/OpenPoseDemo.exe --video &outvideo_path \
            # --write_json &output_snippets_dir
            shutil.rmtree(output_snippets_dir, ignore_errors=True)
            os.makedirs(output_snippets_dir)
            os.system(command_line)

            # pack openpose ouputs
            video = utils.video.get_video_frames(outvideo_path)

            height, width, _ = video[0].shape

            # 这里可以修改label, label_index
            video_info = utils.openpose.json_pack(
                output_snippets_dir, video_name, width, height, label_name, label_no)

            if not os.path.exists(output_sequence_dir):
                os.makedirs(output_sequence_dir)

            with open(output_sequence_path, 'w') as outfile:
                json.dump(video_info, outfile)
            if len(video_info['data']) == 0:
                print('{} Can not find pose estimation results.'.format(file_name))
                return
            else:
                print('{} pose estimation complete.'.format(file_name))

if __name__ == '__main__':
    p = PreProcess()
    p.start()

第四步：手动将生成出来的视频的json文件，照数量（7：3）的比例分到两个文件夹当中，分别做训练集（train）和验证集(val），并生成kinetics_train_label.json和kinetics_val_label.json，代码如下：

##	json2kinetics.py
import json
import os

if __name__ == '__main__':
    train_json_path = 'D:/VIDEOS/mydata/kinetics-skeleton/kinetics_train'    # './mydata/kinetics-skeleton/kinetics_train'
    val_json_path = 'D:/VIDEOS/mydata/kinetics-skeleton/kinetics_val'  # './mydata/kinetics-skeleton/kinetics_val'

    output_train_json_path = 'D:/VIDEOS/mydata/kinetics-skeleton/kinetics_train_label.json'
    output_val_json_path = 'D:/VIDEOS/mydata/kinetics-skeleton/kinetics_val_label.json'

    train_json_names = os.listdir(train_json_path)
    val_json_names = os.listdir(val_json_path)

    train_label_json = dict()
    val_label_json = dict()


    for file_name in train_json_names:
        name = os.path.splitext(file_name)[0]
        json_file_path = '{}/{}'.format(train_json_path, file_name)
        # D:/VIDEOS/mydata/kinetics-skeleton/kinetics_train/{file_name}
        json_file = json.load(open(json_file_path))

        file_label = dict()
        if len(json_file['data']) == 0:
            file_label['has_skeleton'] = False
        else:
            file_label['has_skeleton'] = True
        file_label['label'] = json_file['label']
        file_label['label_index'] = json_file['label_index']

        train_label_json['{}'.format(name)] = file_label

        print('{} success'.format(file_name))

    with open(output_train_json_path, 'w') as outfile:
        json.dump(train_label_json, outfile)

    for file_name in val_json_names:
        name = file_name.split('.')[0]
        json_file_path = '{}/{}'.format(val_json_path, file_name)
        json_file = json.load(open(json_file_path))

        file_label = dict()
        if len(json_file['data']) == 0:
            file_label['has_skeleton'] = False
        else:
            file_label['has_skeleton'] = True
        file_label['label'] = json_file['label']
        file_label['label_index'] = json_file['label_index']

        val_label_json['{}'.format(name)] = file_label

        print('{} success'.format(file_name))

    with open(output_val_json_path, 'w') as outfile:
        json.dump(val_label_json, outfile)

最终生成的样子如下：

然后，就可以送到st-gcn模型里面继续了。