Project that will analyze and get the regex the user from a video. It allows:

• Use a local video file or a video from YouTube.
• Apply any regex to the OCR extracted text.
• Save the results and frames detected in a folder.
• Select how frequently the frames will be analyzed.
• Set the quality of the video if it is from YouTube. By default it will be 1080p60.

• Python 3.9 or higher
• Tesseract installed and added to PATH
• Cuda capable GPU (optional)

1. Clone the repository
2. Install the requirements
3. Run the program with

python .\videoOCR.py -v <video file or yotube uri> -r <regex you want to apply> -o <output folder> -s <every how many frames it will check> -q <quality of the video>