I am so pleased to see we performed so well in Predicting Media Memorability at MultimediaEval Benchmarking.
We improved state-of-the-art and scored a best-in-class value of 0.528 at short-term memorability. Super interesting challenge and very well organized!
Check out my code at https://github.com/dazcona/memorability, our slides and our paper
This is super ... we now use machine learning to predict short-term memorability (several hours) of video clips based on content, composition, aesthetics, emotions, etc. We're now in a position to start asking WHY a video is more or less memorable than another @insight_centre
— Alan Smeaton (@asmeaton) November 1, 2019
Pleased to see we performed so well in Predicting Media Memorability at @MultimediaEval Benchmarking, improving SOTA and scoring a best-in-class value of 0.528 at short-term memorability. Super interesting challenge and well organized! Check out my code https://t.co/XF8uSbhf5p pic.twitter.com/JMBxyUyuQt
— David Azcona (@dazconap) October 30, 2019