Medium Memorability
Predicting media memorability focuses on automatically determining how well people will remember a video clip, a crucial area for improving media design and retrieval. Current research employs diverse approaches, including the fusion of multiple machine learning algorithms (like those based on Particle Swarm Optimization or neural networks such as CNNs and DenseNets) analyzing visual, textual, and auditory features, and even incorporating EEG data to understand neural responses to media. These efforts aim to improve the accuracy of memorability prediction, offering valuable insights for fields ranging from advertising and entertainment to education and archival management.
Papers
Overview of the EEG Pilot Subtask at MediaEval 2021: Predicting Media Memorability
Lorin Sweeney, Ana Matran-Fernandez, Sebastian Halder, Alba G. Seco de Herrera, Alan Smeaton, Graham Healy
Predicting Media Memorability: Comparing Visual, Textual and Auditory Features
Lorin Sweeney, Graham Healy, Alan F. Smeaton