MyLibrary2 のバックアップ(No.11)

私的AI研究会 >MyLibrary2

Python 私的汎用ライブラリ２ †

　適宜再利用できそうなプログラムをライブラリ化するその２

Python 私的汎用ライブラリ２
参考資料

※ 最終更新:2024/06/27　

↑

画像編集パッケージ †

↑

動画編集「my_videotool.py」 †

　「matplotlib.animation」を利用して動画編集をする

関数メソッド (パラメータ詳細はソースコード参照)

機能	戻り値	関数
静止画のリサイズ	image	image_resize(image, size=(256, 256))
動画のリサイズ	video	video_resize(video, size=(256, 256))
静止画と動画 2枚の連結	animation	img_movie3x1(image, video, video2=None, interval=30, deley=0, pixel_size=(256,256), dpi=64)
3枚の動画フレームを連結	animation	merge3x1_ex(listpath, interval=30, deley=0, pixel_size=(256,256), dpi=64):
3枚の動画フレームを連結	animation	merge3x1(source, interval=30, deley=0, pixel_size=(256,256), dpi=64)
4枚の動画フレームを連結	animation	merge2x2(source, interval=30, deley=0, pixel_size=(256,256), dpi=64)
4枚の動画フレームを連結	animation	merge2x2_ex(listpath, interval=30, deley=0, pixel_size=(256,256), dpi=64)
6枚の動画フレームを連結	animation	merge3x2_ex(listpath, interval=30, deley=0, pixel_size=(256,256), dpi=64)
6枚の動画フレームを連結	animation	merge3x2(source, interval=30, deley=0, pixel_size=(256,256), dpi=64)
結果動画の表示と保存（終了処理）	-	save_video(ani, save_path, dispf=False)
動画ファイルを読み出す	video, fps	read_video(filepath))
音声トラックを追加	-	add_audio(source_path, save_path)
リストファイルの読み込み	filelist	read_listfile(filepath)

モジュール・テスト「my_videotool.py」

・コマンドパラメータ

パラメータ	テスト内容
0	静止画と動画 2枚の連結
1	6枚の動画フレームを連結（規定値）
2	4枚の動画フレームを連結（規定値）
3	3枚の動画フレームを連結（規定値）
4	音声トラックを追加
5	リストファイルの読み込み
10	リストファイルから 6枚の動画フレームを連結
20	リストファイルから 4枚の動画フレームを連結
30	リストファイルから 3枚の動画フレームを連結
40	音声トラックを追加

$ python my_videotool.py <option:0/1/2/3/4/5>
$ python my_videotool.py <option:10/20/30/40> <listfile> <savefile>

・テスト・コマンド実行ログ

(py38_learn) python my_videotool.py 0
 Source file → './mylib_test/filelist.txt'
   image file     → './mylib_test/04.jpg'
   driving file   → './mylib_test/2.mp4'
   generated file → './mylib_test/04_output.mp4'
 Saving mpeg file → 'test_anim.mp4'
 Saving gif  file → 'test_anim.gif'

(py38_learn) python my_videotool.py 1
 Source file → './mylib_test/filelist_1.txt'
 Saving mpeg file → 'test_anim1.mp4'
 Saving gif  file → 'test_anim1.gif'

(py38_learn) python my_videotool.py 2
 Source file → './mylib_test/filelist_2.txt'
 Saving mpeg file → 'test_anim2.mp4'
 Saving gif  file → 'test_anim2.gif'

(py38_learn) python my_videotool.py 3
 Source file → './mylib_test/filelist_3.txt'
 Saving mpeg file → 'test_anim3.mp4'
 Saving gif  file → 'test_anim3.gif'

(py38_learn) python my_videotool.py 4
 Source file → './mylib_test/01_video_256.mp4'
 Saving file → 'h1_13_out.mp4'

(py38_learn) python my_videotool.py 5
 Source file → './mylib_test/filelist_1.txt'
 get data: ['./mylib_test/2.mp4', './mylib_test/mon_output.mp4', './mylib_test/statue-02_output.mp4', './mylib_test/04_output.mp4', './mylib_test/pic3_output.mp4', './mylib_test/f_tsuchiya_1_output.mp4']

モジュール・ソースコード

▼「my_videotool.py」

# -*- coding: utf-8 -*-
##------------------------------------------
##   My Library Video Tool  Ver 0.01
##      「matplotlib」を利用した動画編集
##
##               2024.06.14 Masahiro Izutsu
##------------------------------------------
## my_videotool.py

# Color Escape Code
GREEN = '\033[1;32m'
RED = '\033[1;31m'
NOCOLOR = '\033[0m'
YELLOW = '\033[1;33m'

import imageio.v2 as imageio
import numpy as np
import matplotlib.pyplot as plt
import matplotlib.animation as animation
from skimage.transform import resize
import ffmpeg
import tempfile
from shutil import copyfileobj

# 静止画のリサイズ *
def image_resize(image, size = (256, 256)):
    image = resize(image, size)[..., :3]
    return image

# 動画のリサイズ *
def video_resize(video, size = (256, 256)):
    video = [resize(frame, size)[..., :3] for frame in video]
    return video

# 静止画 と 動画 2枚 の連結
def img_movie3x1(image, video, video2 = None, interval = 30, deley = 0, pixel_size = (256,256), dpi = 64):
    # ピクセル → インチ変換
    x_inch = pixel_size[0] / dpi
    y_inch = pixel_size[1] / dpi

    # リサイズ
    image = image_resize(image, pixel_size)
    video = video_resize(video, pixel_size)
    if video2 is not None:
        video2 = video_resize(video2, pixel_size)

    fig = plt.figure(figsize = (x_inch * 2 + x_inch * (video2 is not None), y_inch), dpi = dpi)
    fig.subplots_adjust(left=0, right=1, bottom=0, top=1)
    ims = []
    for i in range(len(video)):
        cols = [image]
        cols.append(video[i])
        if video2 is not None:
            cols.append(video2[i])
        im = plt.imshow(np.concatenate(cols, axis=1), animated = True)
        plt.axis('off')
        ims.append([im])

    ani = animation.ArtistAnimation(fig, ims, interval=interval + 3.4)      # フレームレート補正
    return ani

# 3枚の動画フレームを連結
def merge3x1_ex(listpath, interval = 30, deley = 0, pixel_size = (256,256), dpi = 64):
    return merge3x1(read_listfile(listpath))

def merge3x1(source, interval = 30, deley = 0, pixel_size = (256,256), dpi = 64):
    # ピクセル → インチ変換
    x_inch = pixel_size[0] / dpi
    y_inch = pixel_size[1] / dpi

    result = []
    for file_path in source:
        video = imageio.mimread(file_path, memtest = False)
        result.append(video_resize(video, pixel_size))

    fig = plt.figure(figsize = (x_inch * 3, y_inch), dpi = 64)
    fig.subplots_adjust(left=0, right=1, bottom=0, top=1)
    ims =[]
    for i in range(len(result[0])):
        x = np.concatenate([result[0][i], result[1][i], result[2][i]],1)
        z = np.concatenate([x])
        im = plt.imshow(z, animated=True)
        plt.axis('off')
        ims.append([im])

    ani = animation.ArtistAnimation(fig, ims, interval=interval + 3.36)     # フレームレート補正
    return ani

# 4枚の動画フレームを連結
def merge2x2_ex(listpath, interval = 30, deley = 0, pixel_size = (256,256), dpi = 64):
    return merge2x2(read_listfile(listpath))

def merge2x2(source, interval = 30, deley = 0, pixel_size = (256,256), dpi = 64):
    # ピクセル → インチ変換
    x_inch = pixel_size[0] / dpi
    y_inch = pixel_size[1] / dpi

    result = []
    for file_path in source:
        video = imageio.mimread(file_path, memtest = False)
        result.append(video_resize(video, pixel_size))

    fig = plt.figure(figsize = (x_inch * 2, y_inch * 2), dpi = 64)
    fig.subplots_adjust(left=0, right=1, bottom=0, top=1)
    ims =[]
    for i in range(len(result[0])):
        x = np.concatenate([result[0][i], result[1][i]],1)
        y = np.concatenate([result[2][i], result[3][i]],1)
        z = np.concatenate([x, y])
        im = plt.imshow(z, animated=True)
        plt.axis('off')
        ims.append([im])

    ani = animation.ArtistAnimation(fig, ims, interval=interval + 3.36)     # フレームレート補正
    return ani

# 6枚の動画フレームを連結
def merge3x2_ex(listpath, interval = 30, deley = 0, pixel_size = (256,256), dpi = 64):
    return merge3x2(read_listfile(listpath))

def merge3x2(source, interval = 30, deley = 0, pixel_size = (256,256), dpi = 64):
    # ピクセル → インチ変換
    x_inch = pixel_size[0] / dpi
    y_inch = pixel_size[1] / dpi

    result = []
    for file_path in source:
        video = imageio.mimread(file_path, memtest = False)
        result.append(video_resize(video, pixel_size))

    fig = plt.figure(figsize = (x_inch * 3, y_inch * 2), dpi = 64)
    fig.subplots_adjust(left=0, right=1, bottom=0, top=1)
    ims =[]
    for i in range(len(result[0])):
        x = np.concatenate([result[0][i], result[1][i], result[2][i]],1)
        y = np.concatenate([result[3][i], result[4][i], result[5][i]],1)
        z = np.concatenate([x, y])
        im = plt.imshow(z, animated=True)
        plt.axis('off')
        ims.append([im])

    ani = animation.ArtistAnimation(fig, ims, interval=interval + 3.36)     # フレームレート補正
    return ani


# 結果動画の表示と保存（終了処理）
def save_video(ani, save_path, dispf = False):
    if len(save_path) > 0:
        ani.save(save_path)
    if dispf:
        plt.show()
    plt.close()
    return

# 動画ファイルを読み出す
#   戻り値: video... 動画配列
#           fps      速度（fps）
def read_video(filepath):
    reader = imageio.get_reader(filepath)
    fps = reader.get_meta_data()['fps']
    video = []
    try:
        for im in reader:
            video.append(im)
    except RuntimeError:
        pass
    reader.close()

    return video, fps

# 音声トラックを追加
def add_audio(source_path, save_path):
    try:
        with tempfile.TemporaryDirectory() as str_temp_dir:
            tmpfile = f"{str_temp_dir}/temp.mp4"
            ffmpeg.output(ffmpeg.input(save_path).video, ffmpeg.input(source_path).audio, tmpfile, c='copy').run(quiet=True)
            with open(save_path, 'wb') as result:
                with open(tmpfile, 'rb') as output:
                    copyfileobj(output, result)
    except ffmpeg.Error:
        print(RED + f"Failed to copy audio: no audio track or the audio format is invalid." + NOCOLOR)

# リストファイルの読み込み
#   戻り値: filelist... ファイルリスト
def read_listfile(filepath):

    try:
        f = open(filepath, 'r')
        list = f.readlines()
        filelist = [item[:-1] for item in list]

    except Exception as e:
        print(RED + f'{e}' + NOCOLOR)
        filelist = []

    return filelist

#-----Test routine-----
# $ python my_videotool.py <option:0/1/2/3/4/5>
# $ python my_videotool.py <option:10/20/30/40> <listfile> <savefile>
#
if __name__ == "__main__":
    import os, sys

    args = sys.argv
    sel = '0' if len(args) <= 1 else args[1]

    if sel == '0':                                      # 静止画 と 動画 2枚 の連結
        source_path = './mylib_test/filelist.txt'
        save_path1 = 'test_anim.mp4'
        save_path2 = 'test_anim.gif'

        source_list = read_listfile(source_path)
        image_path = source_list[0]
        driving_path = source_list[1]
        generated_path = source_list[2]

        print(f" Source file → '{source_path}'")
        print(f"   image file     → '{image_path}'")
        print(f"   driving file   → '{driving_path}'")
        print(f"   generated file → '{generated_path}'")
        print(f" Saving mpeg file → '{save_path1}'")
        print(f" Saving gif  file → '{save_path2}'")

        source_image = imageio.imread(image_path)
        driving_video = imageio.mimread(driving_path, memtest = False)
        generated_video = imageio.mimread(generated_path, memtest = False)
        ani = img_movie3x1(source_image, driving_video, generated_video)
        save_video(ani, save_path1, True)
        save_video(ani, save_path2)

    elif sel == '1':                                    # 3枚の動画フレームを連結（規定値）
        source_path = './mylib_test/filelist_1.txt'
        save_path1 = 'test_anim1.mp4'
        save_path2 = 'test_anim1.gif'

        print(f" Source file → '{source_path}'")
        print(f" Saving mpeg file → '{save_path1}'")
        print(f" Saving gif  file → '{save_path2}'")

        ani = merge3x2_ex(source_path)
        save_video(ani, save_path1, True)
        save_video(ani, save_path2)

    elif sel == '2':                                    # 4枚の動画フレームを連結（規定値）
        source_path = './mylib_test/filelist_2.txt'
        save_path1 = 'test_anim2.mp4'
        save_path2 = 'test_anim2.gif'

        print(f" Source file → '{source_path}'")
        print(f" Saving mpeg file → '{save_path1}'")
        print(f" Saving gif  file → '{save_path2}'")

        ani = merge2x2_ex(source_path)
        save_video(ani, save_path1, True)
        save_video(ani, save_path2)

    elif sel == '3':                                    # 3枚の動画フレームを連結（規定値）
        source_path = './mylib_test/filelist_3.txt'
        save_path1 = 'test_anim3.mp4'
        save_path2 = 'test_anim3.gif'

        print(f" Source file → '{source_path}'")
        print(f" Saving mpeg file → '{save_path1}'")
        print(f" Saving gif  file → '{save_path2}'")

        ani = merge3x1_ex(source_path)
        save_video(ani, save_path1, True)
        save_video(ani, save_path2)

    elif sel == '4':
        source_path = './mylib_test/01_video_256.mp4'
        save_path = 'h1_13_out.mp4'

        print(f" Source file → '{source_path}'")
        print(f" Saving file → '{save_path}'")

        add_audio(source_path, save_path)

    elif sel == '5':
        source_path = './mylib_test/filelist_1.txt'
        print(f" Source file → '{source_path}'")

        files = read_listfile('./mylib_test/filelist_1.txt')
        print(f" get data: {files}")

    else:
        if len(args) == 4:
            source_path = args[2]
            save_path = args[3]
            print(f" Source file → '{source_path}'")
            print(f" Saving file → '{save_path}'")

            if sel == '10':                             # 6枚の動画フレームを連結
                ani = merge3x2_ex(source_path)
                save_video(ani, save_path, True)

            elif sel == '20':                           # 4枚の動画フレームを連結
                ani = merge2x2_ex(source_path)
                save_video(ani, save_path, True)

            elif sel == '30':                           # 3枚の動画フレームを連結
                ani = merge3x1_ex(source_path)
                save_video(ani, save_path, True)

            elif sel == '40':                           # 音声トラックを追加
                add_audio(source_path, save_path)

↑

画像編集「my_imagetool.py」 †

　「opencv」を利用した画像編集

関数メソッド (パラメータ詳細はソースコード参照)

機能	戻り値	関数
アスペクト比を固定して画像をリサイズ	image	frame_resize(image, maxsize)
画像を正方形にする（長いほうの辺で）	image	frame_square(image, color=(255, 255, 255))
結果の表示と保存（終了処理）	-	image_disp(image, winname='ImageTool', dispf=True, save_path='')
画像の種類を判別する（imghdr 非推奨対応）	None/True/False	is_pict(filename)
イメージを表示	-	image2disp(filepath='cam', winname='', savepath='', width_f=False, loop_f=False)

モジュール・テスト「my_imagetool.py」

・コマンドパラメータ

パラメータ	テスト内容
0	アスペクト比を固定して画像をリサイズ
1	画像を正方形にする（長いほうの辺で）
2	結果の表示と保存（終了処理）
3	イメージを表示（カメラ入力）
4	イメージを表示（動画/静止画）

$ python my_imagetool.py {0/1/...}

・テスト・コマンド

(py38_learn) python my_imagetool.py 0
...../mylib_test/04.jpg → ./04_out.jpg

(py38_learn) python my_imagetool.py 1
...../mylib_test/fashion003.png → ./fashion003_out.png

(py38_learn) python my_imagetool.py 2

(py38_learn) python my_imagetool.py 3

(py38_learn) python my_imagetool.py 4

モジュール・ソースコード

▼「my_imagetool.py」

# -*- coding: utf-8 -*-
##------------------------------------------
##   My Library Image Tool  Ver 0.02
##
##               2024.06.14 Masahiro Izutsu
##------------------------------------------
## my_imagetool.py
##      Ver 0.02    2024/06/27  動画表示にループ再生フラグを追加

# Color Escape Code
GREEN = '\033[1;32m'
RED = '\033[1;31m'
NOCOLOR = '\033[0m'
YELLOW = '\033[1;33m'
CYAN = '\033[1;36m'

# 定数定義
WINDOW_WIDTH = 640

import os
import cv2
import my_winstat
import numpy as np

# アスペクト比を固定して画像をリサイズ
def frame_resize(image, maxsize):
    if maxsize > 10:
        img_h, img_w = image.shape[:2]
        if (img_w > img_h):
            if (img_w > maxsize):
                height = round(img_h * (maxsize / img_w))
                image = cv2.resize(image, dsize = (maxsize, height))
        else:
            if (img_h > maxsize):
                width =  round(img_w * (maxsize / img_h))
                image = cv2.resize(image, dsize = (width, maxsize))
    return image

# 画像を正方形にする（長いほうの辺で）
def frame_square(image, color = (255, 255, 255)):
    img_h, img_w = image.shape[:2]
    x0 = 0
    y0 = 0
    x1 = 0
    y1 = 0
    if img_h > img_w:
        size = img_h
        x0 = int((size - img_w) / 2)
        x1 = x0 + img_w
        y1 = size
    else:
        size = img_w
        y0 = int((size - img_h) / 2)
        y1 = y0 + img_h
        x1 = size

    # 白ベースの画像を生成
    dist = np.array([size, size, 3])                    # 縦×横 3チャンネル
    img = np.full(dist, color, dtype=np.uint8)
    img[y0:y1, x0:x1] = image                          # 中央に貼り付け
    return img

## 結果の表示と保存（終了処理）
def image_disp(image, winname = 'ImageTool', dispf = True, save_path = ''):
    if len(save_path) > 0:
        cv2.imwrite(save_path, image)
    if dispf:
        cv2.namedWindow(winname, flags = cv2.WINDOW_AUTOSIZE | cv2.WINDOW_GUI_NORMAL) 
        while True:
            cv2.imshow(winname, image)
            key = cv2.waitKey(10)

            if key == 27 or key == 113:                 # 'esc' or 'q'
                cv2.destroyWindow(winname)
                break

            if not my_winstat._is_visible(winname):
                break
        return

# 画像の種類を判別する（imghdr 非推奨対応）
#   戻り値: 'jeg''png'... 画像ファイル
#           'None'        画像ファイル以外 (動画ファイル)
#           'NotFound'    ファイルが存在しない
def is_pict(filename):
    if not os.path.isfile(filename):
        return 'NotFound'

    types = ['.bmp','.png','.jpg','.jpeg','.JPG','.tif']
    for ss in types:
        if filename.endswith(ss):
            return ss
    return 'None'

# イメージを表示
#   filepath:    Absolute path to image file or cam/cam0/cam1.. for camera stream.
def image2disp(filepath = 'cam', winname = '', savepath = '', width_f = False, loop_f = False):
    if filepath.find('cam') == 0 and len(filepath) < 5:
        filepath = 0 if filepath == 'cam' else int(filepath[3])
        isstream = True
    else:
        filetype = is_pict(filepath)
        isstream = filetype == 'None'
        if (filetype == 'NotFound'):
            print(RED + "input file Not found." + NOCOLOR)
            quit()

    # 入力準備
    if (isstream):
        # カメラ 
        cap = cv2.VideoCapture(filepath)
        ret, frame = cap.read()
        loopflg = cap.isOpened()
        if not loopflg:
            print(RED + "Unable to read the camera." + NOCOLOR)
    else:
        # 画像ファイル読み込み
        frame = cv2.imread(filepath)
        if frame is None:
            print(RED + "\nUnable to read the input." + NOCOLOR)
            quit()

        # アスペクト比を固定してリサイズ
        img_h, img_w = frame.shape[:2]
        if width_f and img_w > WINDOW_WIDTH:
            height = round(img_h * (WINDOW_WIDTH / img_w))
            frame = cv2.resize(frame, dsize = (WINDOW_WIDTH, height))
        loopflg = True   # 1回ループ

    # 処理結果の記録 step1
    if len(savepath) > 0:
        if isstream:
            fps = int(cap.get(cv2.CAP_PROP_FPS))
            out_w = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
            out_h = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
            fourcc = cv2.VideoWriter_fourcc('m', 'p', '4', 'v')
            outvideo = cv2.VideoWriter(outpath, fourcc, fps, (out_w, out_h))

    # メインループ 
    while (loopflg):
        if frame is None:
            print(RED + "\nUnable to read the input." + NOCOLOR)
            quit()

        # 画像表示 
        window_name = winname + "  (hit 'q' or 'esc' key to exit)"
        cv2.namedWindow(window_name, flags=cv2.WINDOW_AUTOSIZE | cv2.WINDOW_GUI_NORMAL) 
        cv2.imshow(window_name, frame)

        # 処理結果の記録 step2
        if len(savepath) > 0:
            if (isstream):
                outvideo.write(frame)
            else:
                cv2.imwrite(outpath, frame)

        # 何らかのキーが押されたら終了 
        breakflg = False
        while(True):
            key = cv2.waitKey(30)
            if key == 27 or key == 113:                 # 'esc' or 'q'
                breakflg = True
                cv2.destroyWindow(window_name)
                break
            if not my_winstat._is_visible(window_name): # 'Close' button
                breakflg = True
                break
            if (isstream):
                break

        if ((breakflg == False) and isstream):
            # 次のフレームを読み出す
            ret, frame = cap.read()
            if ret == False:
                if (loop_f):                            # 2024/06/27  動画表示にループ再生フラグを追加
                    #最初のフレームに戻る
                    cap.set(cv2.CAP_PROP_POS_FRAMES, 0)
                    ret, frame = cap.read()
                else:
                    cv2.destroyWindow(window_name)
                    break
            loopflg = cap.isOpened()
        else:
            loopflg = False

    # 終了処理 
    if (isstream):
        cap.release()

        # 処理結果の記録 step3
        if len(savepath) > 0:
            if (isstream):
                outvideo.release()

    return

#-----Test routine-----
# $ python my_imagetool.py {option:0/1/2/3...}
#
if __name__ == "__main__":
    import os, sys
    import my_dialog

    args = sys.argv
    sel = '0' if len(args) <= 1 else args[1]
    filename = my_dialog.select_image_file() if sel == '0' or sel == '1' else ''
    if len(filename) > 0:
        root, ext = os.path.splitext(filename)
        savename = root + '_out' + ext
        base_dir_pair = os.path.split(savename)
        savename = './' + base_dir_pair[1]
        print(f'{filename} → {savename}')

        # frame_resize()
        if sel == '0':
            img = cv2.imread(filename)
            img = frame_resize(img, 256)
            winname = 'ImageTool frame_resize()'
            image_disp(img, winname, True, savename)

        # frame_square()
        elif sel == '1':
            img = cv2.imread(filename)
            img = frame_square(img)
            winname = 'ImageTool frame_square()'
            image_disp(img, winname, True, savename)

    elif sel == '2':
        filepath = 'mylib_test/04.jpg'
        img = cv2.imread(filepath)
        winname = 'ImageTool image_disp()'
        image_disp(img, winname)

    elif sel == '3':
        filepath = 'cam0'
        winname = 'ImageTool image_disp()'
        image2disp(filepath, winname)

    elif sel == '4':
        filename = my_dialog.select_image_movie_file()
        if len(filename) > 0:
            winname = 'ImageTool image_disp()'
            image2disp(filename, winname, loop_f = True)

    cv2.destroyAllWindows()

↑

動画編集 Tips †

　「ffmpeg」実行例は「workspace_pylearn/learn」ディレクトリでの実行を想定~

↑

動画ファイルを正方形にする †

「ffmpeg」コマンドを使用する（例：入力動画の 720x940 を 940x940 にする余白は白）

・「workspace_pylearn/learn/」下記のコマンドを実行する

ffmpeg -i "fashion01.mp4" -vf "yadif=deint=interlaced, scale=w=trunc(ih*dar/2)*2:h=trunc(ih/2)*2, setsar=1/1, scale=w=940:h=940:force_original_aspect_ratio=1, pad=w=940:h=940:x=(ow-iw)/2:y=(oh-ih)/2:color=#ffffff" -pix_fmt yuv420p "fashion01_out.mp4"

▼　- log -

(py38_learn) ffmpeg -i "fashion01.mp4" -vf "yadif=deint=interlaced, scale=w=trunc(ih*dar/2)*2:h=trunc(ih/2)*2, setsar=1/1, scale=w=940:h=940:force_original_aspect_ratio=1, pad=w=940:h=940:x=(ow-iw)/2:y=(oh-ih)/2:color=#ffffff" -pix_fmt yuv420p "fashion01_out.mp4"
ffmpeg version 4.2.2 Copyright (c) 2000-2019 the FFmpeg developers
  built with gcc 9.2.1 (GCC) 20200122
  configuration: --disable-static --enable-shared --enable-gpl --enable-version3 --enable-sdl2 --enable-fontconfig --enable-gnutls --enable-iconv --enable-libass --enable-libdav1d --enable-libbluray --enable-libfreetype --enable-libmp3lame --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libopenjpeg --enable-libopus --enable-libshine --enable-libsnappy --enable-libsoxr --enable-libtheora --enable-libtwolame --enable-libvpx --enable-libwavpack --enable-libwebp --enable-libx264 --enable-libx265 --enable-libxml2 --enable-libzimg --enable-lzma --enable-zlib --enable-gmp --enable-libvidstab --enable-libvorbis --enable-libvo-amrwbenc --enable-libmysofa --enable-libspeex --enable-libxvid --enable-libaom --enable-libmfx --enable-amf --enable-ffnvcodec --enable-cuvid --enable-d3d11va --enable-nvenc --enable-nvdec --enable-dxva2 --enable-avisynth --enable-libopenmpt
  libavutil      56. 31.100 / 56. 31.100
  libavcodec     58. 54.100 / 58. 54.100
  libavformat    58. 29.100 / 58. 29.100
  libavdevice    58.  8.100 / 58.  8.100
  libavfilter     7. 57.100 /  7. 57.100
  libswscale      5.  5.100 /  5.  5.100
  libswresample   3.  5.100 /  3.  5.100
  libpostproc    55.  5.100 / 55.  5.100
Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'fashion01.mp4':
  Metadata:
    major_brand     : isom
    minor_version   : 512
    compatible_brands: isomiso2avc1mp41
    encoder         : Lavf57.71.100
  Duration: 00:00:13.18, start: 0.000000, bitrate: 666 kb/s
    Stream #0:0(eng): Video: h264 (Constrained Baseline) (avc1 / 0x31637661), yuv420p(tv, bt709), 720x940 [SAR 1:1 DAR 36:47], 664 kb/s, 29.97 fps, 29.97 tbr, 30k tbn, 59.94 tbc (default)
    Metadata:
      handler_name    : VideoHandler
Stream mapping:
  Stream #0:0 -> #0:0 (h264 (native) -> h264 (libx264))
Press [q] to stop, [?] for help
[libx264 @ 0000018ed6367bc0] using SAR=1/1
[libx264 @ 0000018ed6367bc0] using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2 AVX FMA3 BMI2 AVX2
[libx264 @ 0000018ed6367bc0] profile High, level 3.1, 4:2:0, 8-bit
[libx264 @ 0000018ed6367bc0] 264 - core 159 - H.264/MPEG-4 AVC codec - Copyleft 2003-2019 - http://www.videolan.org/x264.html - options: cabac=1 ref=3 deblock=1:0:0 analyse=0x3:0x113 me=hex subme=7 psy=1 psy_rd=1.00:0.00 mixed_ref=1 me_range=16 chroma_me=1 trellis=1 8x8dct=1 cqm=0 deadzone=21,11 fast_pskip=1 chroma_qp_offset=-2 threads=29 lookahead_threads=4 sliced_threads=0 nr=0 decimate=1 interlaced=0 bluray_compat=0 constrained_intra=0 bframes=3 b_pyramid=2 b_adapt=1 b_bias=0 direct=1 weightb=1 open_gop=0 weightp=2 keyint=250 keyint_min=25 scenecut=40 intra_refresh=0 rc_lookahead=40 rc=crf mbtree=1 crf=23.0 qcomp=0.60 qpmin=0 qpmax=69 qpstep=4 ip_ratio=1.40 aq=1:1.00
Output #0, mp4, to 'fashion01_out.mp4':
  Metadata:
    major_brand     : isom
    minor_version   : 512
    compatible_brands: isomiso2avc1mp41
    encoder         : Lavf58.29.100
    Stream #0:0(eng): Video: h264 (libx264) (avc1 / 0x31637661), yuv420p, 940x940 [SAR 1:1 DAR 1:1], q=-1--1, 29.97 fps, 30k tbn, 29.97 tbc (default)
    Metadata:
      handler_name    : VideoHandler
      encoder         : Lavc58.54.100 libx264
    Side data:
      cpb: bitrate max/min/avg: 0/0/0 buffer size: 0 vbv_delay: -1
frame=  395 fps=0.0 q=-1.0 Lsize=    1403kB time=00:00:13.07 bitrate= 878.5kbits/s speed=26.9x
video:1397kB audio:0kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 0.390577%
[libx264 @ 0000018ed6367bc0] frame I:2     Avg QP:18.76  size: 23849
[libx264 @ 0000018ed6367bc0] frame P:101   Avg QP:23.73  size:  8531
[libx264 @ 0000018ed6367bc0] frame B:292   Avg QP:27.30  size:  1783
[libx264 @ 0000018ed6367bc0] consecutive B-frames:  1.0%  0.5%  2.3% 96.2%
[libx264 @ 0000018ed6367bc0] mb I  I16..4: 13.2% 74.8% 11.9%
[libx264 @ 0000018ed6367bc0] mb P  I16..4:  0.3%  1.6%  0.8%  P16..4:  6.4%  5.0%  3.0%  0.0%  0.0%    skip:82.9%
[libx264 @ 0000018ed6367bc0] mb B  I16..4:  0.0%  0.0%  0.0%  B16..8:  9.6%  2.7%  0.7%  direct: 0.4%  skip:86.6%  L0:45.2% L1:42.0% BI:12.9%
[libx264 @ 0000018ed6367bc0] 8x8 transform intra:65.6% inter:45.9%
[libx264 @ 0000018ed6367bc0] coded y,uvDC,uvAC intra: 40.9% 33.8% 13.0% inter: 3.4% 1.6% 0.0%
[libx264 @ 0000018ed6367bc0] i16 v,h,dc,p: 87%  5%  2%  7%
[libx264 @ 0000018ed6367bc0] i8 v,h,dc,ddl,ddr,vr,hd,vl,hu: 39%  5% 38%  2%  3%  4%  2%  4%  3%
[libx264 @ 0000018ed6367bc0] i4 v,h,dc,ddl,ddr,vr,hd,vl,hu: 33% 10% 20%  5%  6%  9%  5%  7%  5%
[libx264 @ 0000018ed6367bc0] i8c dc,h,v,p: 68%  7% 22%  3%
[libx264 @ 0000018ed6367bc0] Weighted P-Frames: Y:0.0% UV:0.0%
[libx264 @ 0000018ed6367bc0] ref P L0: 70.8% 12.6% 13.4%  3.2%
[libx264 @ 0000018ed6367bc0] ref B L0: 95.1%  4.3%  0.6%
[libx264 @ 0000018ed6367bc0] ref B L1: 98.0%  2.0%
[libx264 @ 0000018ed6367bc0] kb/s:868.01

・参考サイト：

どんな解像度・縦横比の動画も一発で指定のサイズにするffmpegコマンド

↑

動画ファイルの一部を切り取る †

「ffmpeg」コマンドを使用する（例：入力動画の 450,80 座標から 340x340 を切り取る）

・「workspace_pylearn/learn/」下記のコマンドを実行する

ffmpeg -i 01.mp4 -ss 00:00:00.00 -t 00:00:03 -filter:v "crop=340:340:450:80" -async 1 01_video_340.mp4

▼　- log -

(py38_learn) ffmpeg -i 01.mp4 -ss 00:00:00.00 -t 00:00:03 -filter:v "crop=340:340:450:80" -async 1 01_video_340.mp4
ffmpeg version 4.2.2 Copyright (c) 2000-2019 the FFmpeg developers
  built with gcc 9.2.1 (GCC) 20200122
  configuration: --disable-static --enable-shared --enable-gpl --enable-version3 --enable-sdl2 --enable-fontconfig --enable-gnutls --enable-iconv --enable-libass --enable-libdav1d --enable-libbluray --enable-libfreetype --enable-libmp3lame --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libopenjpeg --enable-libopus --enable-libshine --enable-libsnappy --enable-libsoxr --enable-libtheora --enable-libtwolame --enable-libvpx --enable-libwavpack --enable-libwebp --enable-libx264 --enable-libx265 --enable-libxml2 --enable-libzimg --enable-lzma --enable-zlib --enable-gmp --enable-libvidstab --enable-libvorbis --enable-libvo-amrwbenc --enable-libmysofa --enable-libspeex --enable-libxvid --enable-libaom --enable-libmfx --enable-amf --enable-ffnvcodec --enable-cuvid --enable-d3d11va --enable-nvenc --enable-nvdec --enable-dxva2 --enable-avisynth --enable-libopenmpt
  libavutil      56. 31.100 / 56. 31.100
  libavcodec     58. 54.100 / 58. 54.100
  libavformat    58. 29.100 / 58. 29.100
  libavdevice    58.  8.100 / 58.  8.100
  libavfilter     7. 57.100 /  7. 57.100
  libswscale      5.  5.100 /  5.  5.100
  libswresample   3.  5.100 /  3.  5.100
  libpostproc    55.  5.100 / 55.  5.100
Input #0, mov,mp4,m4a,3gp,3g2,mj2, from '01.mp4':
  Metadata:
    major_brand     : mp42
    minor_version   : 1
    compatible_brands: isommp41mp42
    creation_time   : 2022-01-28T13:08:43.000000Z
  Duration: 00:00:02.94, start: 0.000000, bitrate: 945 kb/s
    Stream #0:0(und): Video: h264 (High) (avc1 / 0x31637661), yuv420p(tv, bt709), 1280x720 [SAR 1:1 DAR 16:9], 796 kb/s, 29.84 fps, 29.97 tbr, 30k tbn, 59.94 tbc (default)
    Metadata:
      creation_time   : 2022-01-28T13:08:43.000000Z
      handler_name    : Core Media Video
    Stream #0:1(und): Audio: aac (LC) (mp4a / 0x6134706D), 44100 Hz, stereo, fltp, 127 kb/s (default)
    Metadata:
      creation_time   : 2022-01-28T13:08:43.000000Z
      handler_name    : Core Media Audio
Stream mapping:
  Stream #0:0 -> #0:0 (h264 (native) -> h264 (libx264))
  Stream #0:1 -> #0:1 (aac (native) -> aac (native))
Press [q] to stop, [?] for help
-async is forwarded to lavfi similarly to -af aresample=async=1:min_hard_comp=0.100000:first_pts=0.
[libx264 @ 0000018188207ac0] using SAR=1/1
[libx264 @ 0000018188207ac0] using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2 AVX FMA3 BMI2 AVX2
[libx264 @ 0000018188207ac0] profile High, level 2.1, 4:2:0, 8-bit
[libx264 @ 0000018188207ac0] 264 - core 159 - H.264/MPEG-4 AVC codec - Copyleft 2003-2019 - http://www.videolan.org/x264.html - options: cabac=1 ref=3 deblock=1:0:0 analyse=0x3:0x113 me=hex subme=7 psy=1 psy_rd=1.00:0.00 mixed_ref=1 me_range=16 chroma_me=1 trellis=1 8x8dct=1 cqm=0 deadzone=21,11 fast_pskip=1 chroma_qp_offset=-2 threads=11 lookahead_threads=1 sliced_threads=0 nr=0 decimate=1 interlaced=0 bluray_compat=0 constrained_intra=0 bframes=3 b_pyramid=2 b_adapt=1 b_bias=0 direct=1 weightb=1 open_gop=0 weightp=2 keyint=250 keyint_min=25 scenecut=40 intra_refresh=0 rc_lookahead=40 rc=crf mbtree=1 crf=23.0 qcomp=0.60 qpmin=0 qpmax=69 qpstep=4 ip_ratio=1.40 aq=1:1.00
Output #0, mp4, to '01_video_340.mp4':
  Metadata:
    major_brand     : mp42
    minor_version   : 1
    compatible_brands: isommp41mp42
    encoder         : Lavf58.29.100
    Stream #0:0(und): Video: h264 (libx264) (avc1 / 0x31637661), yuv420p, 340x340 [SAR 1:1 DAR 1:1], q=-1--1, 29.97 fps, 30k tbn, 29.97 tbc (default)
    Metadata:
      creation_time   : 2022-01-28T13:08:43.000000Z
      handler_name    : Core Media Video
      encoder         : Lavc58.54.100 libx264
    Side data:
      cpb: bitrate max/min/avg: 0/0/0 buffer size: 0 vbv_delay: -1
    Stream #0:1(und): Audio: aac (LC) (mp4a / 0x6134706D), 44100 Hz, stereo, fltp, 128 kb/s (default)
    Metadata:
      creation_time   : 2022-01-28T13:08:43.000000Z
      handler_name    : Core Media Audio
      encoder         : Lavc58.54.100 aac
frame=   88 fps=0.0 q=-1.0 Lsize=     144kB time=00:00:02.94 bitrate= 400.2kbits/s speed=25.9x
video:93kB audio:46kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 3.205245%
[libx264 @ 0000018188207ac0] frame I:1     Avg QP:21.58  size:  7014
[libx264 @ 0000018188207ac0] frame P:24    Avg QP:22.92  size:  2542
[libx264 @ 0000018188207ac0] frame B:63    Avg QP:26.62  size:   423
[libx264 @ 0000018188207ac0] consecutive B-frames:  1.1%  9.1%  3.4% 86.4%
[libx264 @ 0000018188207ac0] mb I  I16..4: 13.2% 72.1% 14.7%
[libx264 @ 0000018188207ac0] mb P  I16..4:  1.8%  5.9%  0.8%  P16..4: 38.3% 17.5%  7.8%  0.0%  0.0%    skip:27.8%
[libx264 @ 0000018188207ac0] mb B  I16..4:  0.1%  0.1%  0.0%  B16..8: 36.3%  3.0%  0.3%  direct: 0.6%  skip:59.6%  L0:39.1% L1:53.1% BI: 7.9%
[libx264 @ 0000018188207ac0] 8x8 transform intra:69.5% inter:78.6%
[libx264 @ 0000018188207ac0] coded y,uvDC,uvAC intra: 48.7% 66.9% 19.8% inter: 9.0% 8.2% 0.2%
[libx264 @ 0000018188207ac0] i16 v,h,dc,p: 20% 39% 24% 17%
[libx264 @ 0000018188207ac0] i8 v,h,dc,ddl,ddr,vr,hd,vl,hu: 17% 14% 30%  5%  7% 10%  5%  8%  4%
[libx264 @ 0000018188207ac0] i4 v,h,dc,ddl,ddr,vr,hd,vl,hu: 34% 11% 12%  6% 11% 10%  5%  8%  3%
[libx264 @ 0000018188207ac0] i8c dc,h,v,p: 48% 22% 22%  7%
[libx264 @ 0000018188207ac0] Weighted P-Frames: Y:0.0% UV:0.0%
[libx264 @ 0000018188207ac0] ref P L0: 65.2% 16.3% 14.3%  4.2%
[libx264 @ 0000018188207ac0] ref B L0: 93.4%  5.5%  1.1%
[libx264 @ 0000018188207ac0] ref B L1: 97.3%  2.7%
[libx264 @ 0000018188207ac0] kb/s:257.87
[aac @ 0000018188276440] Qavg: 189.911

・引数

-i：動画ファイルのパス
-ss：動画のクロップスタート時刻(時間：分：秒：1/100秒)
-t：クロップ動画の長さ(時間：分：秒)
-filter:v “crop=XX:XX:XX:XX”：(クロップ画像の横サイズ:クロップ画像の縦サイズ:クロップ画像の左上角の横位置:クロップ画像の左上角の縦位置)
＊クロップ画像の左上角の横位置・縦位置は、動画の左上角を(0,0)として計算した位置
-async 1：クロップ画像保存ファイルパス

その他の例

・入力動画の 780,50 座標から 600x600 を切り取るコマンド

ffmpeg -i 07.mkv -ss 00:08:57.50 -t 00:00:08 -filter:v "crop=600:600:760:50" -async 1 hinton.mp4

・入力動画の 550,80 座標から 200x200 を切り取るコマンド

ffmpeg -i news_x.mp4 -ss 00:00:00.00 -t 00:00:10 -filter:v "crop=200:200:550:80" -async 1 news01.mp4

↑

動画に音を追加する †

「ffmpeg」コマンドを使用する（例：動画ファイル「outvideo6_17.mp4」音声ファイル「hi_audio.mp3」）
・「workspace_pylearn/learn/」で下記のコマンドを実行する

ffmpeg -i outvideo6_17.mp4 -i hi_audio.mp3 -c:v copy -c:a mp3 -map 0:v:0 -map 1:a:0 outvideo6_17_oudio.mp4

▼　- log -

(py38_learn) ffmpeg -i outvideo6_17.mp4 -i hi_audio.mp3 -c:v copy -c:a mp3 -map 0:v:0 -map 1:a:0 outvideo6_17_oudio.mp4
ffmpeg version 4.2.2 Copyright (c) 2000-2019 the FFmpeg developers
  built with gcc 9.2.1 (GCC) 20200122
  configuration: --disable-static --enable-shared --enable-gpl --enable-version3 --enable-sdl2 --enable-fontconfig --enable-gnutls --enable-iconv --enable-libass --enable-libdav1d --enable-libbluray --enable-libfreetype --enable-libmp3lame --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libopenjpeg --enable-libopus --enable-libshine --enable-libsnappy --enable-libsoxr --enable-libtheora --enable-libtwolame --enable-libvpx --enable-libwavpack --enable-libwebp --enable-libx264 --enable-libx265 --enable-libxml2 --enable-libzimg --enable-lzma --enable-zlib --enable-gmp --enable-libvidstab --enable-libvorbis --enable-libvo-amrwbenc --enable-libmysofa --enable-libspeex --enable-libxvid --enable-libaom --enable-libmfx --enable-amf --enable-ffnvcodec --enable-cuvid --enable-d3d11va --enable-nvenc --enable-nvdec --enable-dxva2 --enable-avisynth --enable-libopenmpt
  libavutil      56. 31.100 / 56. 31.100
  libavcodec     58. 54.100 / 58. 54.100
  libavformat    58. 29.100 / 58. 29.100
  libavdevice    58.  8.100 / 58.  8.100
  libavfilter     7. 57.100 /  7. 57.100
  libswscale      5.  5.100 /  5.  5.100
  libswresample   3.  5.100 /  3.  5.100
  libpostproc    55.  5.100 / 55.  5.100
Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'outvideo6_17.mp4':
  Metadata:
    major_brand     : isom
    minor_version   : 512
    compatible_brands: isomiso2avc1mp41
    encoder         : Lavf58.76.100
  Duration: 00:00:02.90, start: 0.000000, bitrate: 999 kb/s
    Stream #0:0(und): Video: h264 (High) (avc1 / 0x31637661), yuv420p, 768x512, 994 kb/s, 30.30 fps, 30.30 tbr, 16k tbn, 60.61 tbc (default)
    Metadata:
      handler_name    : VideoHandler
Input #1, mp3, from 'hi_audio.mp3':
  Metadata:
    encoder         : Lavf58.76.100
  Duration: 00:00:02.95, start: 0.025057, bitrate: 128 kb/s
    Stream #1:0: Audio: mp3, 44100 Hz, stereo, fltp, 128 kb/s
    Metadata:
      encoder         : Lavc58.13
Stream mapping:
  Stream #0:0 -> #0:0 (copy)
  Stream #1:0 -> #0:1 (mp3 (mp3float) -> mp3 (libmp3lame))
Press [q] to stop, [?] for help
Output #0, mp4, to 'outvideo6_17_oudio.mp4':
  Metadata:
    major_brand     : isom
    minor_version   : 512
    compatible_brands: isomiso2avc1mp41
    encoder         : Lavf58.29.100
    Stream #0:0(und): Video: h264 (High) (avc1 / 0x31637661), yuv420p, 768x512, q=2-31, 994 kb/s, 30.30 fps, 30.30 tbr, 16k tbn, 16k tbc (default)
    Metadata:
      handler_name    : VideoHandler
    Stream #0:1: Audio: mp3 (libmp3lame) (mp4a / 0x6134706D), 44100 Hz, stereo, fltp
    Metadata:
      encoder         : Lavc58.54.100 libmp3lame
[libmp3lame @ 0000025df4091f80] Trying to remove 1152 samples, but the queue is empty
frame=   88 fps=0.0 q=-1.0 Lsize=     403kB time=00:00:02.92 bitrate=1127.6kbits/s speed=98.3x
video:353kB audio:46kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 1.002753%

・この方法で合成した mp4 ファイルは Webサーバーによる配信では表示できないことがある

下記 Webサービスの利用が簡単！
→ ビデオに音楽を追加 - 無料でビデオにオーディオを追加

↑

動画のサイズを変更する †

「ffmpeg」コマンドを使用する

・「workspace_pylearn/learn/」で下記のコマンドを実行する（例：入力動画を 256x256 サイズに変換）

ffmpeg -i 01_video_340.mp4 -s 256:256 -q 2 01_video_256.mp4

▼　- log -

(py38_learn) ffmpeg -i 01_video_340.mp4 -s 256:256 -q 2 01_video_256.mp4
ffmpeg version 4.2.2 Copyright (c) 2000-2019 the FFmpeg developers
  built with gcc 9.2.1 (GCC) 20200122
  configuration: --disable-static --enable-shared --enable-gpl --enable-version3 --enable-sdl2 --enable-fontconfig --enable-gnutls --enable-iconv --enable-libass --enable-libdav1d --enable-libbluray --enable-libfreetype --enable-libmp3lame --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libopenjpeg --enable-libopus --enable-libshine --enable-libsnappy --enable-libsoxr --enable-libtheora --enable-libtwolame --enable-libvpx --enable-libwavpack --enable-libwebp --enable-libx264 --enable-libx265 --enable-libxml2 --enable-libzimg --enable-lzma --enable-zlib --enable-gmp --enable-libvidstab --enable-libvorbis --enable-libvo-amrwbenc --enable-libmysofa --enable-libspeex --enable-libxvid --enable-libaom --enable-libmfx --enable-amf --enable-ffnvcodec --enable-cuvid --enable-d3d11va --enable-nvenc --enable-nvdec --enable-dxva2 --enable-avisynth --enable-libopenmpt
  libavutil      56. 31.100 / 56. 31.100
  libavcodec     58. 54.100 / 58. 54.100
  libavformat    58. 29.100 / 58. 29.100
  libavdevice    58.  8.100 / 58.  8.100
  libavfilter     7. 57.100 /  7. 57.100
  libswscale      5.  5.100 /  5.  5.100
  libswresample   3.  5.100 /  3.  5.100
  libpostproc    55.  5.100 / 55.  5.100
Input #0, mov,mp4,m4a,3gp,3g2,mj2, from '01_video_340.mp4':
  Metadata:
    major_brand     : isom
    minor_version   : 512
    compatible_brands: isomiso2avc1mp41
    encoder         : Lavf58.29.100
  Duration: 00:00:02.97, start: 0.000000, bitrate: 396 kb/s
    Stream #0:0(und): Video: h264 (High) (avc1 / 0x31637661), yuv420p, 340x340 [SAR 1:1 DAR 1:1], 259 kb/s, 29.97 fps, 29.97 tbr, 30k tbn, 59.94 tbc (default)
    Metadata:
      handler_name    : Core Media Video
    Stream #0:1(und): Audio: aac (LC) (mp4a / 0x6134706D), 44100 Hz, stereo, fltp, 128 kb/s (default)
    Metadata:
      handler_name    : Core Media Audio
Stream mapping:
  Stream #0:0 -> #0:0 (h264 (native) -> h264 (libx264))
  Stream #0:1 -> #0:1 (aac (native) -> aac (native))
Press [q] to stop, [?] for help
[libx264 @ 00000266fd448740] -qscale is ignored, -crf is recommended.
[libx264 @ 00000266fd448740] using SAR=1/1
[libx264 @ 00000266fd448740] using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2 AVX FMA3 BMI2 AVX2
[libx264 @ 00000266fd448740] profile High, level 1.3, 4:2:0, 8-bit
[libx264 @ 00000266fd448740] 264 - core 159 - H.264/MPEG-4 AVC codec - Copyleft 2003-2019 - http://www.videolan.org/x264.html - options: cabac=1 ref=3 deblock=1:0:0 analyse=0x3:0x113 me=hex subme=7 psy=1 psy_rd=1.00:0.00 mixed_ref=1 me_range=16 chroma_me=1 trellis=1 8x8dct=1 cqm=0 deadzone=21,11 fast_pskip=1 chroma_qp_offset=-2 threads=8 lookahead_threads=1 sliced_threads=0 nr=0 decimate=1 interlaced=0 bluray_compat=0 constrained_intra=0 bframes=3 b_pyramid=2 b_adapt=1 b_bias=0 direct=1 weightb=1 open_gop=0 weightp=2 keyint=250 keyint_min=25 scenecut=40 intra_refresh=0 rc_lookahead=40 rc=crf mbtree=1 crf=23.0 qcomp=0.60 qpmin=0 qpmax=69 qpstep=4 ip_ratio=1.40 aq=1:1.00
Output #0, mp4, to '01_video_256.mp4':
  Metadata:
    major_brand     : isom
    minor_version   : 512
    compatible_brands: isomiso2avc1mp41
    encoder         : Lavf58.29.100
    Stream #0:0(und): Video: h264 (libx264) (avc1 / 0x31637661), yuv420p, 256x256 [SAR 1:1 DAR 1:1], q=-1--1, 29.97 fps, 30k tbn, 29.97 tbc (default)
    Metadata:
      handler_name    : Core Media Video
      encoder         : Lavc58.54.100 libx264
    Side data:
      cpb: bitrate max/min/avg: 0/0/0 buffer size: 0 vbv_delay: -1
    Stream #0:1(und): Audio: aac (LC) (mp4a / 0x6134706D), 44100 Hz, stereo, fltp, 128 kb/s (default)
    Metadata:
      handler_name    : Core Media Audio
      encoder         : Lavc58.54.100 aac
frame=   88 fps=0.0 q=-1.0 Lsize=     122kB time=00:00:02.94 bitrate= 339.8kbits/s speed=  40x
video:56kB audio:62kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 3.780259%
[libx264 @ 00000266fd448740] frame I:1     Avg QP:21.81  size:  4763
[libx264 @ 00000266fd448740] frame P:27    Avg QP:23.32  size:  1479
[libx264 @ 00000266fd448740] frame B:60    Avg QP:25.73  size:   203
[libx264 @ 00000266fd448740] consecutive B-frames:  2.3% 18.2%  6.8% 72.7%
[libx264 @ 00000266fd448740] mb I  I16..4: 18.8% 68.8% 12.5%
[libx264 @ 00000266fd448740] mb P  I16..4:  1.1%  2.3%  0.4%  P16..4: 37.9% 24.7% 11.1%  0.0%  0.0%    skip:22.5%
[libx264 @ 00000266fd448740] mb B  I16..4:  0.0%  0.1%  0.0%  B16..8: 35.2%  3.7%  0.3%  direct: 0.3%  skip:60.5%  L0:36.0% L1:55.3% BI: 8.7%
[libx264 @ 00000266fd448740] 8x8 transform intra:64.7% inter:72.3%
[libx264 @ 00000266fd448740] coded y,uvDC,uvAC intra: 51.0% 72.2% 30.2% inter: 10.0% 8.9% 0.3%
[libx264 @ 00000266fd448740] i16 v,h,dc,p: 23% 35% 18% 24%
[libx264 @ 00000266fd448740] i8 v,h,dc,ddl,ddr,vr,hd,vl,hu: 13% 12% 27%  6%  9% 12%  6% 11%  5%
[libx264 @ 00000266fd448740] i4 v,h,dc,ddl,ddr,vr,hd,vl,hu: 20% 20% 10%  9% 11% 10%  8%  6%  5%
[libx264 @ 00000266fd448740] i8c dc,h,v,p: 56% 20% 18%  6%
[libx264 @ 00000266fd448740] Weighted P-Frames: Y:0.0% UV:0.0%
[libx264 @ 00000266fd448740] ref P L0: 66.0% 16.2% 13.3%  4.5%
[libx264 @ 00000266fd448740] ref B L0: 94.6%  4.8%  0.6%
[libx264 @ 00000266fd448740] ref B L1: 98.0%  2.0%
[libx264 @ 00000266fd448740] kb/s:155.03
[aac @ 00000266ff1fdcc0] Qavg: 236.000

↑

YouTube 動画をダウンロードする †

パッケージをインストール
```
(py38_gan) PS > pip install yt-dlp
```

モジュール・ソースコード

▼「ytb_down.py」

# YouTube ダウンロード
#   2023.11.18
# ytb_down.py
# ffmpeg yt-dlp
#    test:    opt = ['https://www.youtube.com/watch?v=7G0ovtPqHnI']
#
# 【簡単便利】pythonからYouTubeを高画質でdownloadするぞ！by Windows
# https://resanaplaza.com/2023/04/02/%E3%80%90%E7%B0%A1%E5%8D%98%E4%BE%BF%E5%88%A9%E3%80%91python%E3%81%8B%E3%82%89youtube%E3%82%92%E9%AB%98%E7%94%BB%E8%B3%AA%E3%81%A7download%E3%81%99%E3%82%8B%E3%81%9E%EF%BC%81by-windows/

from yt_dlp import YoutubeDL

def ytb_download(option):
    ydl = YoutubeDL()
    result = ydl.download(option)
    return result

if __name__ == "__main__":
    import sys

    opt = []
    args = sys.argv
    for n in range(len(args)):
        print(n, args[n])
        if n >= 1:
            opt.append(args[n])

    if len(opt) == 0:
        print('Please input potion parameter !!')

    print(opt)

    result = ytb_download(opt)

ダウンロードする YouTube 動画のURL を右クリックでコピーしておく

コマンドラインにペーストして実行する（ローカルマシン py38_learn 仮想環境）

python ytb_down.py 'https://youtu.be/cZM3o5KWLKQ'

▼　- log -

(py38_learn) python ytb_down.py 'https://youtu.be/cZM3o5KWLKQ'
0 ytb_down.py
1 https://youtu.be/cZM3o5KWLKQ
['https://youtu.be/cZM3o5KWLKQ']
[youtube] Extracting URL: https://youtu.be/cZM3o5KWLKQ
[youtube] cZM3o5KWLKQ: Downloading webpage
[youtube] cZM3o5KWLKQ: Downloading ios player API JSON
[youtube] cZM3o5KWLKQ: Downloading android player API JSON
WARNING: [youtube] YouTube said: ERROR - Precondition check failed.
WARNING: [youtube] HTTP Error 400: Bad Request. Retrying (1/3)...
[youtube] cZM3o5KWLKQ: Downloading android player API JSON
WARNING: [youtube] YouTube said: ERROR - Precondition check failed.
WARNING: [youtube] HTTP Error 400: Bad Request. Retrying (2/3)...
[youtube] cZM3o5KWLKQ: Downloading android player API JSON
WARNING: [youtube] YouTube said: ERROR - Precondition check failed.
WARNING: [youtube] HTTP Error 400: Bad Request. Retrying (3/3)...
[youtube] cZM3o5KWLKQ: Downloading android player API JSON
WARNING: [youtube] YouTube said: ERROR - Precondition check failed.
WARNING: [youtube] Unable to download API page: HTTP Error 400: Bad Request (caused by <HTTPError 400: Bad Request>); please report this issue on  https://github.com/yt-dlp/yt-dlp/issues?q= , filling out the appropriate issue template. Confirm you are on the latest version using  yt-dlp -U
[youtube] cZM3o5KWLKQ: Downloading m3u8 information
[info] cZM3o5KWLKQ: Downloading 1 format(s): 248+251
[download] Destination: 花王 セグレタ 「肌のように、髪を愛そう。」篇　30秒 CM 柴咲コウ [cZM3o5KWLKQ].f248.webm
[download] 100% of    4.96MiB in 00:00:00 at 34.11MiB/s
[download] Destination: 花王 セグレタ 「肌のように、髪を愛そう。」篇　30秒 CM 柴咲コウ [cZM3o5KWLKQ].f251.webm
[download] 100% of  495.41KiB in 00:00:00 at 10.58MiB/s
[Merger] Merging formats into "花王 セグレタ 「肌のように、髪を愛そう。」篇　30秒 CM 柴咲コウ [cZM3o5KWLKQ].webm"
Deleting original file 花王 セグレタ 「肌のように、髪を愛そう。」篇　30秒 CM 柴咲コウ [cZM3o5KWLKQ].f251.webm (pass -k to keep)
Deleting original file 花王 セグレタ 「肌のように、髪を愛そう。」篇　30秒 CM 柴咲コウ [cZM3o5KWLKQ].f248.webm (pass -k to keep)

↑

対処した問題点とエラー詳細 †

↑

「ffmpeg」パッケージのインストール †

パッケージを調べる

(py38_learn) pip install ffmpeg==
ERROR: Could not find a version that satisfies the requirement ffmpeg== (from versions: 1.1.0, 1.2.0, 1.2.1, 1.3, 1.4)
ERROR: No matching distribution found for ffmpeg==

(py38_learn) conda search ffmpeg
Loading channels: done
# Name                       Version           Build  Channel
ffmpeg                           4.2      ha925a31_0  pkgs/main
ffmpeg                           4.2      hc56fc5f_0  pkgs/main
ffmpeg                         4.2.2      he774522_0  pkgs/main

新しい方をインストールする

conda install ffmpeg

pip install ffmpeg-python

ツールパッケージをインストール
```
conda search imageio-ffmpeg -c conda-forge
```

↑

「scikit-learn」パッケージのインストール †

インストールしていない場合

pip install scikit-learn

▼　- log -

(py38_learn) pip install scikit-learn
Collecting scikit-learn
  Downloading scikit_learn-1.3.2-cp38-cp38-win_amd64.whl.metadata (11 kB)
Requirement already satisfied: numpy<2.0,>=1.17.3 in c:\users\izuts\anaconda3\envs\py38_learn\lib\site-packages (from scikit-learn) (1.24.4)
Requirement already satisfied: scipy>=1.5.0 in c:\users\izuts\anaconda3\envs\py38_learn\lib\site-packages (from scikit-learn) (1.9.0)
Requirement already satisfied: joblib>=1.1.1 in c:\users\izuts\anaconda3\envs\py38_learn\lib\site-packages (from scikit-learn) (1.3.2)
Collecting threadpoolctl>=2.0.0 (from scikit-learn)
  Downloading threadpoolctl-3.5.0-py3-none-any.whl.metadata (13 kB)
Downloading scikit_learn-1.3.2-cp38-cp38-win_amd64.whl (9.3 MB)
   ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 9.3/9.3 MB 49.5 MB/s eta 0:00:00
Downloading threadpoolctl-3.5.0-py3-none-any.whl (18 kB)
Installing collected packages: threadpoolctl, scikit-learn
Successfully installed scikit-learn-1.3.2 threadpoolctl-3.5.0

↑

更新履歴 †

2024/06/14 初版

↑

参考資料 †

Python

matplotlib

ffmpeg

other
- python動画編集モジュールmoviepy 解説
- Pythonでmp4の動画から音声のみを抽出したmp3ファイルを作成する