Our paper “VIDEM: VIDeo Effectiveness and Memorability Dataset” by Rukiye Savran Kızıltepe, Sohail Sahab, Rodrigo Valladares Santana, Faiyaz Doctor, Kate Paterson, David Hunstone, and Alba G. Seco de Herrera has been accepted for presentation at IWANN 2025.

VIDEM introduces a large-scale dataset designed to support research on video effectiveness, memorability, and brand recognition. It includes 424 YouTube videos annotated by over 1,400 participants, using the Video Effectiveness Measurement Tool (VEMT) developed with advanced crowdsourcing and quality control. The dataset is now available as part of the MediaEval 2025 Benchmark: https://multimediaeval.github.io/editions/2025/tasks/memorability/.

We hope VIDEM will enable new insights into video memorability, branding impact, and marketing analytics, and foster further research in multimedia information retrieval.