File size: 3,969 Bytes
5767e72 | 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 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 | import json
import os
import argparse
import sys
def check_num_videos(json_path, video_dir):
"""
Checks if the total number of video files matches the expected count
based on the input JSON filename.
json file for i2v task is assumed to contain the string 'i2v'
"""
# Logic preserved from original script:
# distinct expected counts for i2v vs t2v based on filename
if 'i2v' in os.path.basename(json_path).lower():
expected_count = 5590
else:
expected_count = 4720
print(f"--- Checking counts for: {os.path.basename(json_path)} ---")
# Check if directory exists
if not os.path.isdir(video_dir):
print(f"! Error: Directory '{video_dir}' not found.")
return False
# Count files ending in .mp4
files = [f for f in os.listdir(video_dir) if f.lower().endswith('.mp4')]
actual_count = len(files)
# Compare and Report
print(f"Expected videos: {expected_count}")
print(f"Found videos: {actual_count}")
if actual_count == expected_count:
print("OK! check_num_videos: SUCCESS. Count matches.")
return True
else:
diff = expected_count - actual_count
if diff > 0:
print(f"! check_num_videos: FAIL. Missing {diff} videos.")
else:
print(f"! check_num_videos: FAIL. Found {abs(diff)} extra videos.")
return False
def filter_missing_videos(json_path, video_dir, output_path):
"""
Identifies specifically which entries from the JSON are missing
corresponding video files in the directory.
"""
if not os.path.exists(json_path):
print(f"! Error: Input JSON '{json_path}' not found.")
return
try:
with open(json_path, 'r') as f:
data = json.load(f)
except json.JSONDecodeError:
print(f"! Error: Failed to decode JSON from '{json_path}'.")
return
missing_entries = []
print(f"\n--- Scanning {len(data)} entries against '{video_dir}' ---")
for entry in data:
# Adjust key if your json uses something other than 'image_name'
# Fallback to 'prompt' or other keys if necessary, but using original logic:
image_name = entry.get('image_name', '')
if not image_name:
continue
# Logic: Convert 'xxx.jpg' -> 'xxx' -> 'xxx-0.mp4'
base_name = os.path.splitext(image_name)[0]
expected_video_name = f"{base_name}-0.mp4"
expected_video_path = os.path.join(video_dir, expected_video_name)
# If video file is NOT found, add the whole entry to our list
if not os.path.exists(expected_video_path):
missing_entries.append(entry)
# Save the missing entries to a new file
if missing_entries:
with open(output_path, 'w') as f:
json.dump(missing_entries, f, indent=4)
print(f"! Found {len(missing_entries)} missing videos.")
print(f"Data for missing entries saved to: {output_path}")
else:
print("OK! All videos exist! No missing entries found.")
if __name__ == "__main__":
parser = argparse.ArgumentParser(
description="Check for missing sampled videos against a VBench JSON reference."
)
# Positional arguments
parser.add_argument(
"json_path",
type=str,
help="Path to the original JSON file (e.g., vbench2_i2v_full_info.json)"
)
parser.add_argument(
"video_dir",
type=str,
help="Path to the folder containing sampled .mp4 files"
)
# Optional argument
parser.add_argument(
"--output",
type=str,
default="missing_entries.json",
help="Path to save the resulting JSON of missing entries (default: missing_entries.json)"
)
args = parser.parse_args()
# Run the checks
check_num_videos(args.json_path, args.video_dir)
filter_missing_videos(args.json_path, args.video_dir, args.output) |