Trim your video files for training ML Models. While training a model for use in A.I, we needed a way to prepare our training data. We had a lot of video files and our use case didn't require long video files so this script was written to solve that. We just feed in the video files and it trims it down to the size we want with all the audio files. A metadata.csv file is also created which contains all the timestamps for each gif produced. So if you are in need of this kind of use case or just wanna create gifs, this is a faily good tool to use.
- Create two folders, Videos and Output.
- In the Output folder, create two more folders, audio and gif
- In the Videos folder, place all the video files you want trimmed.
The trimmed video gifs will be placed in the gif folder, the audio for each gif will be placed in the audio folder. You can change how the duration for which the videos are trimmed by editing the gif_duration parameter of the code.
This script is written using modules from the MoviePy library. MoviePy