Skip to content

Latest commit

 

History

History
72 lines (43 loc) · 3.98 KB

stacked_README.md

File metadata and controls

72 lines (43 loc) · 3.98 KB
Raw

Stacked - Portrait Stacked

This Python script enhances video presentations by combining 2 video sources (example: speaker footage and slide content) into a single, visually appealing portrait video.

This allows content such as Datadog On to be repurposed on social media in a format optimized for mobile viewing, featuring a distinct background color and improved accessibility.

NOTE: It is highly recommended to run the black_bars.py script prior to running this script. This crops the original video containing slides so that the black bars at the top and bottom are removed (if necessary).

WARNING: This process could take some time depending on the length of your original video source. If you are testing things and want to see an output from a shorter video, it is recommended you try out the trimmer.py script to reduce your original video to something much smaller to work with.

Features

  • Video Resizing: Utilizes ffmpeg to resize both video sources (i.e. speaker and slide videos) to a target width while maintaining aspect ratios, ensuring content is visually consistent and optimized for the target resolution.
  • Dynamic Background: Introduces a colored background, customizable to match branding or aesthetic preferences, enhancing the overall look of the combined video.
  • Audio Preservation: Maintains the original audio from the speaker video, ensuring the message is delivered without compromise.
  • Vertical Layout Optimization: Arranges speaker footage and slides in a vertical sequence, making the final video suitable for platforms favoring portrait orientation.

Workflow Overview

  1. Preparation: Resizes the input videos (speaker footage and slides) to a designated width, ensuring compatibility with the target display resolution.

  2. Background Creation: Generates a color clip using a specified color, serving as the backdrop for the combined video.

  3. Vertical Assembly: Stacks the resized speaker video atop the slide video against the colorful background, optimizing the layout for portrait orientation.

  4. Audio Integration: Attaches the original audio from the speaker video to the final composition, preserving the presentation's auditory component.

  5. Export: Outputs the assembled video to a file, ready for distribution or upload to video-sharing platforms.

Requirements

  • Python 3.x
  • moviepy library for video processing in Python.
  • ffmpeg installed and accessible from the command line for video resizing tasks.

Installation

Ensure you have Python 3.x installed on your system. Install the required Python library using pip:

pip install moviepy

Verify that ffmpeg is installed:

ffmpeg -version

If ffmpeg is not installed, download and install it from FFmpeg's official website or through your system's package manager.

Usage Instructions

  1. Customize the script parameters (input_path, output_path, desired_width, background_color, etc.) to suit your video processing needs.

  2. Run the script:

python script_name.py

Example

Given trimmed speaker footage (video/original/speakers.mp4) and corresponding slide content (video/original/slides.mp4), the script generates a single video (video/original/combined_video_portrait.mp4). This video features both elements against a Datadog Purple background, arranged for optimal mobile viewing, with the speaker's original audio.

Customization

  • Background Color: Modify the background_color variable to align with your brand or preferred aesthetic.
  • Target Resolution: Adjust desired_width and desired_height to match your output resolution requirements.
  • Additional Adjustments: Further customize the positioning, scaling, or duration of individual components as needed for your specific project.

This script streamlines the creation of engaging, mobile-friendly video content, making it an invaluable tool for educators, marketers, and content creators aiming to enhance their video presentations.