Textual and visual summary of videos
Budget: $250 – $750 USD
Set up a web interface that takes as input the link of a video from the Shamengo Youtube channel then summarizes it as follows:
1. Analyze the video transcripts then extract the different named entities so that they can answer the 5W1H questions (When, where, Who, What, Why, How)
2. Make the association of these different entities that tell the story of the video to video time codes. Thus each entity corresponds to a sequence of the video
Indeed, the answer to the 5W1H questions given by the extracted entities consists in textual summary of the video, then the mapping of the entities to the time codes of the video provides a visual summary of it.
1. Analyze the video transcripts then extract the different named entities so that they can answer the 5W1H questions (When, where, Who, What, Why, How)
2. Make the association of these different entities that tell the story of the video to video time codes. Thus each entity corresponds to a sequence of the video
Indeed, the answer to the 5W1H questions given by the extracted entities consists in textual summary of the video, then the mapping of the entities to the time codes of the video provides a visual summary of it.