Enregistré dans:
Détails bibliographiques
Auteurs principaux: Kumar, Prajneya, Khandelwal, Eshika, Tapaswi, Makarand, Sreekumar, Vishnu
Format: Preprint
Publié: 2023
Sujets:
Accès en ligne:https://arxiv.org/abs/2311.16484
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866910684120350720
author Kumar, Prajneya
Khandelwal, Eshika
Tapaswi, Makarand
Sreekumar, Vishnu
author_facet Kumar, Prajneya
Khandelwal, Eshika
Tapaswi, Makarand
Sreekumar, Vishnu
contents Understanding what makes a video memorable has important applications in advertising or education technology. Towards this goal, we investigate spatio-temporal attention mechanisms underlying video memorability. Different from previous works that fuse multiple features, we adopt a simple CNN+Transformer architecture that enables analysis of spatio-temporal attention while matching state-of-the-art (SoTA) performance on video memorability prediction. We compare model attention against human gaze fixations collected through a small-scale eye-tracking study where humans perform the video memory task. We uncover the following insights: (i) Quantitative saliency metrics show that our model, trained only to predict a memorability score, exhibits similar spatial attention patterns to human gaze, especially for more memorable videos. (ii) The model assigns greater importance to initial frames in a video, mimicking human attention patterns. (iii) Panoptic segmentation reveals that both (model and humans) assign a greater share of attention to things and less attention to stuff as compared to their occurrence probability.
format Preprint
id arxiv_https___arxiv_org_abs_2311_16484
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Seeing Eye to AI: Comparing Human Gaze and Model Attention in Video Memorability
Kumar, Prajneya
Khandelwal, Eshika
Tapaswi, Makarand
Sreekumar, Vishnu
Computer Vision and Pattern Recognition
Understanding what makes a video memorable has important applications in advertising or education technology. Towards this goal, we investigate spatio-temporal attention mechanisms underlying video memorability. Different from previous works that fuse multiple features, we adopt a simple CNN+Transformer architecture that enables analysis of spatio-temporal attention while matching state-of-the-art (SoTA) performance on video memorability prediction. We compare model attention against human gaze fixations collected through a small-scale eye-tracking study where humans perform the video memory task. We uncover the following insights: (i) Quantitative saliency metrics show that our model, trained only to predict a memorability score, exhibits similar spatial attention patterns to human gaze, especially for more memorable videos. (ii) The model assigns greater importance to initial frames in a video, mimicking human attention patterns. (iii) Panoptic segmentation reveals that both (model and humans) assign a greater share of attention to things and less attention to stuff as compared to their occurrence probability.
title Seeing Eye to AI: Comparing Human Gaze and Model Attention in Video Memorability
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2311.16484