Datasets:
Delete preprocess_videos.py
Browse files- preprocess_videos.py +0 -137
preprocess_videos.py
DELETED
@@ -1,137 +0,0 @@
|
|
1 |
-
import os
|
2 |
-
import sys
|
3 |
-
import subprocess
|
4 |
-
import cv2
|
5 |
-
import numpy as np
|
6 |
-
from decord import VideoReader, cpu
|
7 |
-
from tqdm import tqdm
|
8 |
-
from PIL import Image
|
9 |
-
import argparse
|
10 |
-
import re
|
11 |
-
|
12 |
-
def get_new_height_width(height, width, longest_side_size=512):
|
13 |
-
if height > width:
|
14 |
-
new_height = longest_side_size
|
15 |
-
new_width = int(new_height * width / height) # floor value
|
16 |
-
else:
|
17 |
-
new_width = longest_side_size
|
18 |
-
new_height = int(new_width * height / width) # floor value
|
19 |
-
return new_height, new_width
|
20 |
-
|
21 |
-
def downsample_video(input_video_path, output_video_path):
|
22 |
-
print(f"Downsampling video: {input_video_path} to {output_video_path}")
|
23 |
-
if os.path.exists(output_video_path):
|
24 |
-
print('Downsampled video exists. Exiting now...')
|
25 |
-
return
|
26 |
-
|
27 |
-
try:
|
28 |
-
print("Attempting to use FFmpeg for fast downsampling...")
|
29 |
-
probe_cmd = ["ffprobe", "-v", "error", "-select_streams", "v:0", "-show_entries",
|
30 |
-
"stream=width,height,r_frame_rate,duration", "-of", "csv=p=0", input_video_path]
|
31 |
-
|
32 |
-
try:
|
33 |
-
probe_output = subprocess.check_output(probe_cmd, universal_newlines=True).strip().split(',')
|
34 |
-
width, height, fps_fraction, duration = probe_output
|
35 |
-
original_fps = eval(fps_fraction)
|
36 |
-
width, height = int(width), int(height)
|
37 |
-
new_height, new_width = get_new_height_width(height, width, longest_side_size=512)
|
38 |
-
total_duration = float(duration)
|
39 |
-
except:
|
40 |
-
video_reader = VideoReader(input_video_path, ctx=cpu(0), num_threads=0)
|
41 |
-
original_fps = video_reader.get_avg_fps()
|
42 |
-
first_frame = video_reader[0].asnumpy()
|
43 |
-
height, width, _ = first_frame.shape
|
44 |
-
new_height, new_width = get_new_height_width(height, width, longest_side_size=512)
|
45 |
-
total_duration = len(video_reader) / original_fps
|
46 |
-
del video_reader
|
47 |
-
|
48 |
-
temp_output_path = output_video_path.split('.')[0] + '.temp.mp4'
|
49 |
-
if os.path.exists(temp_output_path):
|
50 |
-
os.remove(temp_output_path)
|
51 |
-
|
52 |
-
ffmpeg_cmd = [
|
53 |
-
"ffmpeg", "-y",
|
54 |
-
"-i", input_video_path,
|
55 |
-
"-vf", f"fps={original_fps},scale={new_width}:{new_height}",
|
56 |
-
"-c:v", "libx264",
|
57 |
-
"-preset", "medium",
|
58 |
-
"-crf", "23",
|
59 |
-
"-g", "60",
|
60 |
-
"-bf", "2",
|
61 |
-
"-an",
|
62 |
-
"-pix_fmt", "yuv420p",
|
63 |
-
temp_output_path
|
64 |
-
]
|
65 |
-
|
66 |
-
# 使用tqdm显示ffmpeg的进度
|
67 |
-
with tqdm(total=total_duration, desc=f"Processing {os.path.basename(input_video_path)}", leave=False) as pbar:
|
68 |
-
process = subprocess.Popen(ffmpeg_cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, universal_newlines=True)
|
69 |
-
for line in process.stderr:
|
70 |
-
# 匹配ffmpeg输出中的时间信息
|
71 |
-
match = re.search(r'time=(\d+:\d+:\d+\.\d+)', line)
|
72 |
-
if match:
|
73 |
-
time_str = match.group(1)
|
74 |
-
h, m, s = map(float, time_str.split(':'))
|
75 |
-
current_time = h * 3600 + m * 60 + s
|
76 |
-
pbar.update(current_time - pbar.n)
|
77 |
-
process.wait()
|
78 |
-
|
79 |
-
if process.returncode == 0:
|
80 |
-
os.replace(temp_output_path, output_video_path)
|
81 |
-
print(f"Successfully downsampled video using FFmpeg")
|
82 |
-
return
|
83 |
-
|
84 |
-
except (subprocess.SubprocessError, FileNotFoundError) as e:
|
85 |
-
print(f"FFmpeg method failed: {str(e)}. Falling back to Python implementation.")
|
86 |
-
if os.path.exists(temp_output_path):
|
87 |
-
os.remove(temp_output_path)
|
88 |
-
raise
|
89 |
-
|
90 |
-
|
91 |
-
def process_video(filename, source_dir, target_dir):
|
92 |
-
if '_' not in filename and filename.endswith(('.mp4', '.avi', '.mov', '.mkv')):
|
93 |
-
input_video_path = os.path.join(source_dir, filename)
|
94 |
-
output_video_path = os.path.join(target_dir, filename)
|
95 |
-
print(f"Processing {filename}...")
|
96 |
-
|
97 |
-
downsample_video(input_video_path, output_video_path)
|
98 |
-
|
99 |
-
def main():
|
100 |
-
parser = argparse.ArgumentParser(description="Downsample videos from a source directory to a target directory.")
|
101 |
-
parser.add_argument('source_dir', type=str, help='The directory containing the source videos.')
|
102 |
-
parser.add_argument('target_dir', type=str, help='The directory to save the downsampled videos.')
|
103 |
-
parser.add_argument('video_ids_file', type=str, help='The file containing video IDs to process.')
|
104 |
-
|
105 |
-
args = parser.parse_args()
|
106 |
-
|
107 |
-
source_dir = args.source_dir
|
108 |
-
target_dir = args.target_dir
|
109 |
-
video_ids_file = args.video_ids_file
|
110 |
-
|
111 |
-
if not os.path.exists(target_dir):
|
112 |
-
os.makedirs(target_dir)
|
113 |
-
|
114 |
-
with open(video_ids_file, 'r') as f:
|
115 |
-
video_ids = {line.strip() for line in f if line.strip()}
|
116 |
-
|
117 |
-
filenames = [f for f in os.listdir(source_dir) if f.split('.')[0] in video_ids and f.endswith(('.mp4', '.avi', '.mov', '.mkv'))]
|
118 |
-
|
119 |
-
print(f"Found {len(filenames)} videos to process.")
|
120 |
-
|
121 |
-
if not filenames:
|
122 |
-
print("No videos found to process. Exiting.")
|
123 |
-
return
|
124 |
-
|
125 |
-
print(f"Processing {len(filenames)} videos sequentially...")
|
126 |
-
with tqdm(total=len(filenames), desc="Overall Progress") as pbar:
|
127 |
-
for filename in filenames:
|
128 |
-
try:
|
129 |
-
process_video(filename, source_dir, target_dir)
|
130 |
-
except Exception as e:
|
131 |
-
print(f"An error occurred during processing {filename}: {e}")
|
132 |
-
pbar.update(1)
|
133 |
-
|
134 |
-
print("Processing completed successfully.")
|
135 |
-
|
136 |
-
if __name__ == "__main__":
|
137 |
-
main()
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|