It is My ongoing Minor-2 project of college. It observed that sometimes we need all the slides used in any video, so i decided to make a Machine Learning model with python which will take video file as input and returns a pdf file contains all the slides used in that particular video,so to do that i used FFMPEG tool to get images from video and further did data cleaning as i only needs those images which are slides (not contains face of lecturer and also removal of redundant images) using face recognition and text extraction from image.Also in addition I generated video transcript of lecture using speech to text (Machine learning).I have learned lot from this project about Machine learning ,FFMPEG and ML algorithms.
Additionally i generated a text file for time stemp of all the slides so that if user want to search perticular slide in whole video she/he can quickly go to that perticular time stamp on video without wasting too much time on linearly search the slide.
Thankyou...