From 39f2cbb8e6adb52c9ff1df7d096083345839945b Mon Sep 17 00:00:00 2001
From: Ameer Azam <30064373+AMEERAZAM08@users.noreply.github.com>
Date: Wed, 9 Jul 2025 12:14:17 +0000
Subject: [PATCH] Added Youtube Video Downloader script and Video Splitter as
 per Mention in Paper

---
 donwload_video.py   | 74 +++++++++++++++++++++++++++++++++++++++++
 split_time_video.py | 80 +++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 154 insertions(+)
 create mode 100644 donwload_video.py
 create mode 100644 split_time_video.py

diff --git a/donwload_video.py b/donwload_video.py
new file mode 100644
index 0000000..73fea1d
--- /dev/null
+++ b/donwload_video.py
@@ -0,0 +1,74 @@
+from pytube import YouTube
+from moviepy.editor import VideoFileClip, AudioFileClip, CompositeAudioClip
+import os 
+
+
+
+def rename_video(video_path, new_name):
+    # Get the directory path and the current filename
+    directory, filename = os.path.split(video_path)
+    
+    # Remove spaces from the new name and keep the file extension
+    new_name = new_name.replace(" ", "_")
+    
+    # Split the filename and extension
+    base_name, extension = os.path.splitext(filename)
+    
+    # Create the new filename with the new name and original extension
+    new_filename = f"{new_name}{extension}"
+    
+    # Create the new path with the new filename
+    new_path = os.path.join(directory, new_filename)
+    
+    # Rename the file
+    os.rename(video_path, new_path)
+
+
+# Read video URLs from the text file
+file_path = "HDTF_dataset/WRA_video_url.txt"
+
+with open(file_path, "r") as file:
+    video_urls = file.readlines()
+
+output_dir = "HDTF/video_dataset/"
+# Iterate through each video URL and download
+for url in video_urls:
+    try:
+        vid_name, vid_url = url.strip().split(" ")
+        # Replace spaces with underscores in the video name
+        vid_name = vid_name.replace(" ", "_")
+        # Download the video
+        
+        yt = YouTube(vid_url.strip())
+        video = yt.streams.filter(progressive=True, file_extension='mp4').order_by('resolution').desc().first()
+        # Create output directory if it doesn't exist
+
+        # os.makedirs(output_dir, exist_ok=True)
+        # Download the video with the modified name
+        video_path_ = video.download(output_dir)
+        rename_video(video_path_,vid_name)
+
+        print(f"Downloaded: {yt.title}")
+    except Exception as e:
+        print(f"Error downloading {url}: {e}")
+
+
+
+
+
+# # Adjust frame rate to 25fps
+# output_path = '/path_to_save_video/adjusted_video.mp4'
+# clip = VideoFileClip(video_path)
+# clip = clip.set_fps(25)
+
+# # Adjust audio to match the new frame rate
+# audio_clip = AudioFileClip(video_path)
+# adjusted_audio = audio_clip.set_fps(25)
+
+# # Composite video with adjusted audio
+# final_clip = clip.set_audio(adjusted_audio)
+
+# # Write the final clip with adjusted frame rate and audio
+# final_clip.write_videofile(output_path)
+
+# print("Video downloaded and frame rate adjusted successfully with matching audio.")
\ No newline at end of file
diff --git a/split_time_video.py b/split_time_video.py
new file mode 100644
index 0000000..73e344c
--- /dev/null
+++ b/split_time_video.py
@@ -0,0 +1,80 @@
+from moviepy.editor import VideoFileClip
+import os
+
+def split_video_with_audio(video_path, time_intervals, output_directory):
+    # Load the video clip
+    video_clip = VideoFileClip(video_path)
+    audio_clip = video_clip.audio
+    
+    # Set the audio for the entire video clip
+    video_clip = video_clip.set_audio(audio_clip)
+    
+    # Iterate over each time interval
+    for i, time_interval in enumerate(time_intervals):
+        # Parse the time interval
+        start_time, end_time = time_interval.split('-')
+        start_minutes, start_seconds = map(int, start_time.split(':'))
+        end_minutes, end_seconds = map(int, end_time.split(':'))
+        
+        # Convert start and end times to seconds
+        start_time_seconds = start_minutes * 60 + start_seconds
+        end_time_seconds = end_minutes * 60 + end_seconds
+        
+        # Extract the clip
+        clip = video_clip.subclip(start_time_seconds, end_time_seconds)
+        
+        # Create the output directory if it doesn't exist
+        os.makedirs(output_directory, exist_ok=True)
+        
+        # Construct the output file path
+        output_filename = os.path.join(output_directory, f"{os.path.splitext(os.path.basename(video_path))[0]}_{i+1}.mp4")
+        
+        clip.write_videofile(output_filename, codec="libx264")
+        
+        # Close the clip
+        clip.close()
+
+    # Close the original video clip
+    video_clip.close()
+
+# Example usage:
+annot_time_split = "HDTF_dataset/WRA_annotion_time.txt"
+
+with open(annot_time_split, "r") as file:
+    annot_time_split_data = file.readlines()
+for a_d in annot_time_split_data:
+    vid_name, interval = a_d.split(" ")[0],a_d.split(" ")[1:]
+    print(vid_name,interval)
+    video_path = f"HDTF/video_dataset/{vid_name}"
+    time_intervals = interval # List of time intervals in the format "start-end"
+    output_directory = "./split_time"  # Output directory where the videos will be saved
+    # exit()
+    try:
+        split_video_with_audio(video_path, time_intervals, output_directory)
+    except  Exception as e :
+        print("Error is getting " ,e)
+
+
+
+# import os
+
+# def write_sorted_video_names_to_txt(directory, output_file):
+#     # Get all files in the directory
+#     files = os.listdir(directory)
+    
+#     # Filter out only the video files
+#     video_files = [file for file in files if file.endswith('.mp4')]
+    
+#     # Sort the video file names
+#     sorted_video_files = sorted(video_files)
+    
+#     # Write sorted video names to the output file
+#     with open(output_file, 'w') as f:
+#         for video_file in sorted_video_files:
+#             f.write(video_file + '\n')
+
+# # Example usage:
+# directory = "HDTF/split_time"
+# output_file = "./training_video_name.txt"
+
+# write_sorted_video_names_to_txt(directory, output_file)
\ No newline at end of file