We present a novel method for saliency prediction that leverages a cognitive model of visual attention as an inductive bias. This approach is in stark contrast to recent purely data-driven saliency models that achieve performance improvements mainly by increased capacity, resulting in high computational costs and the need for large-scale training datasets. We demonstrate that by using a cognitive model, our method achieves competitive performance to the state of the art across several natural image datasets while only requiring a fraction of the parameters. Furthermore, we set the new state of the art for saliency prediction on information visualizations, demonstrating the effectiveness of our approach for cross-domain generalization. We further provide augmented versions of the full MSCOCO dataset with synthetic gaze data using the cognitive model, which we used to pre-train our method. Our results are highly promising and underline the significant potential of bridging between cognitive and data-driven models, potentially also beyond attention.
%0 Conference Paper
%1 sood23_cogsci
%A Sood, Ekta
%A Shi, Lei
%A Bortoletto, Matteo
%A Wang, Yao
%A Müller, Philipp
%A Bulling, Andreas
%B Proc. of the 45th Annual Meeting of the Cognitive Science Society (CogSci)
%D 2023
%K visus:bullinas visus: sfbtrr161 a07 visus:shili 2023 visus:wangyo visus from:christinawarren visus:bortolmo visus:soodea
%P 1--8
%T Improving Neural Saliency Prediction with a Cognitive Model of Human Visual Attention
%X We present a novel method for saliency prediction that leverages a cognitive model of visual attention as an inductive bias. This approach is in stark contrast to recent purely data-driven saliency models that achieve performance improvements mainly by increased capacity, resulting in high computational costs and the need for large-scale training datasets. We demonstrate that by using a cognitive model, our method achieves competitive performance to the state of the art across several natural image datasets while only requiring a fraction of the parameters. Furthermore, we set the new state of the art for saliency prediction on information visualizations, demonstrating the effectiveness of our approach for cross-domain generalization. We further provide augmented versions of the full MSCOCO dataset with synthetic gaze data using the cognitive model, which we used to pre-train our method. Our results are highly promising and underline the significant potential of bridging between cognitive and data-driven models, potentially also beyond attention.
@inproceedings{sood23_cogsci,
abstract = {We present a novel method for saliency prediction that leverages a cognitive model of visual attention as an inductive bias. This approach is in stark contrast to recent purely data-driven saliency models that achieve performance improvements mainly by increased capacity, resulting in high computational costs and the need for large-scale training datasets. We demonstrate that by using a cognitive model, our method achieves competitive performance to the state of the art across several natural image datasets while only requiring a fraction of the parameters. Furthermore, we set the new state of the art for saliency prediction on information visualizations, demonstrating the effectiveness of our approach for cross-domain generalization. We further provide augmented versions of the full MSCOCO dataset with synthetic gaze data using the cognitive model, which we used to pre-train our method. Our results are highly promising and underline the significant potential of bridging between cognitive and data-driven models, potentially also beyond attention. },
added-at = {2023-05-22T07:49:26.000+0200},
author = {Sood, Ekta and Shi, Lei and Bortoletto, Matteo and Wang, Yao and Müller, Philipp and Bulling, Andreas},
biburl = {https://puma.ub.uni-stuttgart.de/bibtex/243490aa9b74f61dbb51f414d0c8bce57/visus},
booktitle = {Proc. of the 45th Annual Meeting of the Cognitive Science Society (CogSci)},
interhash = {9039d069dd9d3f940693c0f44272cb3f},
intrahash = {43490aa9b74f61dbb51f414d0c8bce57},
keywords = {visus:bullinas visus: sfbtrr161 a07 visus:shili 2023 visus:wangyo visus from:christinawarren visus:bortolmo visus:soodea},
month = {7},
pages = {1--8},
timestamp = {2023-05-22T07:49:26.000+0200},
title = {Improving Neural Saliency Prediction with a Cognitive Model of Human Visual Attention},
year = 2023
}