A Survey on Image Memorability Prediction: From Traditional to Deep Learning Models
Souad Lahrache, Rajae El Ouazzani · 2022 2nd International Conference on Innovative Research in Applied Science, Engineering and Technology (IRASET) · 2022
Image memorability prediction is an interesting and challenging task which aims to create a representative model of memorable images, in order to recognize memorable ones from others. Actually, memorability can be valuable in different areas and several applications. In recent years, image memorability has received considerable interest in computer vision and machine learning communities. Different approaches have been proposed which have made a significant impact on memorability analysis performances. Thus, the purpose of this paper is to expose different methods and techniques used to predict and analyze image memorability. Therefore, we expose and compare datasets and evaluation metrics used for performance memorability estimation. Then, we discuss many challenges and opportunities for image memorability prediction improvement, which can provide some future guidelines.