File size: 1,964 Bytes
3bbb319 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 |
from __future__ import unicode_literals
import argparse
from subprocess import call
import cv2
import numpy as np
import os
import shutil
import pandas as pd
from tqdm import tqdm
import time
parser = argparse.ArgumentParser()
parser.add_argument('-base_path', '--base_path', help='base folder path of dataset')
parser.add_argument('-speaker', '--speaker',
help='download videos of a specific speaker {oliver, jon, conan, rock, chemistry, ellen, almaram, angelica, seth, shelly}')
args = parser.parse_args()
BASE_PATH = args.base_path
df = pd.read_csv(os.path.join(BASE_PATH, "videos_links.csv"))
if args.speaker:
df = df[df['speaker'] == args.speaker]
temp_output_path = './tmp2/temp_video.mp4'
for _, row in tqdm(df.iterrows(), total=df.shape[0]):
i, name, link = row
if 'youtube' in link:
try:
output_path = os.path.join(BASE_PATH, row["speaker"], "videos", row["video_fn"])
if not (os.path.exists(os.path.dirname(output_path))):
os.makedirs(os.path.dirname(output_path))
command = 'yt-dlp -o {temp_path} -f mp4 {link}'.format(link=link, temp_path=temp_output_path)
res1 = call(command, shell=True)
cam = cv2.VideoCapture(temp_output_path)
if np.isclose(cam.get(cv2.CAP_PROP_FPS), 29.97, atol=0.03):
cam.release()
shutil.move(temp_output_path, output_path)
else:
res2 = call('ffmpeg -i "%s" -r 30000/1001 -strict -2 "%s" -y' % (temp_output_path, output_path),
shell=True)
except Exception as e:
print (e)
finally:
if os.path.exists(temp_output_path):
os.remove(temp_output_path)
print("Out of a total of %s videos for %s: "%(len(df), args.speaker))
print("Successfully downloaded:")
my_cmd = 'ls ' + os.path.join(BASE_PATH, row["speaker"], "videos") + ' | wc -l'
os.system(my_cmd)
|