Overview / Usage

I declare myself as a media junkie. I decided to develop an algorithm capable of detecting relevance on news and social networking. This is nothing new. The problem with news and media today is that no one likes to read. GIfs are a standard and I decided to start to make gif based on news.

This is my new project over tensorflow. I am using convolution neural network for taxonomy detection and model detection. My plan is to select the most important seconds of a video to generate the gif I like. Then I generate a gif and publish on social media. I use the googe compute engice for video to detect stuff on videos and cut them on the right spot.

This is just a proof of concept but I bealive that with a lot of training. I may realise a browser extension for videos in order to see animated gifs insted of clickbait thumbnails.