MediaEval 2018: Predicting Media Memorability Task
This task addresses the problem of predicting video memorability for researchers and practitioners in multimedia evaluation, but it is incremental as it builds on prior image memorability work by extending it to videos and adding long-term annotations.
The paper introduces the Predicting Media Memorability task for MediaEval 2018, focusing on automatically predicting memorability scores for videos, with a dataset that includes both short-term and long-term annotations.
In this paper, we present the Predicting Media Memorability task, which is proposed as part of the MediaEval 2018 Benchmarking Initiative for Multimedia Evaluation. Participants are expected to design systems that automatically predict memorability scores for videos, which reflect the probability of a video being remembered. In contrast to previous work in image memorability prediction, where memorability was measured a few minutes after memorization, the proposed dataset comes with short-term and long-term memorability annotations. All task characteristics are described, namely: the task's challenges and breakthrough, the released data set and ground truth, the required participant runs and the evaluation metrics.