import os import csv import yt_dlp as youtube_dl import cv2 from moviepy.editor import VideoFileClip def download_and_extract_frames(csv_file, output_base_folder='video_frames'): # Create base output folder if it doesn't exist os.makedirs(output_base_folder, exist_ok=True) # Read the CSV file with open(csv_file, 'r') as file: csv_reader = csv.reader(file) # Skip header if exists next(csv_reader, None) # Process each video for row in csv_reader: try: video_name, video_link = row # Sanitize video name for folder creation safe_video_name = ''.join(c if c.isalnum() or c in ['-', '_'] else '_' for c in video_name) # Setup download options ydl_opts = { 'outtmpl': f'./{safe_video_name}.%(ext)s', 'format': 'bestvideo+bestaudio/best', 'no_color': True, # Disable color output 'no_warnings': True # Suppress warnings } # Download video with youtube_dl.YoutubeDL(ydl_opts) as ydl: ydl.download([video_link]) # Find the downloaded video file downloaded_files = os.listdir('.') video_files = [f for f in downloaded_files if f.startswith(safe_video_name) and (f.endswith('.mp4') or f.endswith('.webm') or f.endswith('.mkv'))] if not video_files: print(f"Error: No video file found for {video_name}") continue # Create folder for frames frame_folder = os.path.join(output_base_folder, safe_video_name) os.makedirs(frame_folder, exist_ok=True) # Process each downloaded video file for video_file in video_files: # Extract frames video = cv2.VideoCapture(video_file) frame_count = 0 while True: ret, frame = video.read() if not ret: break # Save frame frame_path = os.path.join(frame_folder, f'frame_{frame_count:04d}.jpg') cv2.imwrite(frame_path, frame) frame_count += 1 # Release video capture video.release() print(f"Processed {video_name} from {video_file}: {frame_count} frames extracted") except Exception as e: print(f"Error processing {video_name}: {str(e)}") # Continue to next video if one fails continue # Usage if __name__ == "__main__": download_and_extract_frames('videos.csv')