TempSAL -- Uncovering Temporal Information for Deep Saliency Prediction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Aydemir, Bahar, Hoffstetter, Ludo, Zhang, Tong, Salzmann, Mathieu, Süsstrunk, Sabine
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913494543106048
author Aydemir, Bahar
Hoffstetter, Ludo
Zhang, Tong
Salzmann, Mathieu
Süsstrunk, Sabine
author_facet Aydemir, Bahar
Hoffstetter, Ludo
Zhang, Tong
Salzmann, Mathieu
Süsstrunk, Sabine
contents Deep saliency prediction algorithms complement the object recognition features, they typically rely on additional information, such as scene context, semantic relationships, gaze direction, and object dissimilarity. However, none of these models consider the temporal nature of gaze shifts during image observation. We introduce a novel saliency prediction model that learns to output saliency maps in sequential time intervals by exploiting human temporal attention patterns. Our approach locally modulates the saliency predictions by combining the learned temporal maps. Our experiments show that our method outperforms the state-of-the-art models, including a multi-duration saliency model, on the SALICON benchmark. Our code will be publicly available on GitHub.
format Preprint
id arxiv_https___arxiv_org_abs_2301_02315
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle TempSAL -- Uncovering Temporal Information for Deep Saliency Prediction
Aydemir, Bahar
Hoffstetter, Ludo
Zhang, Tong
Salzmann, Mathieu
Süsstrunk, Sabine
Computer Vision and Pattern Recognition
Deep saliency prediction algorithms complement the object recognition features, they typically rely on additional information, such as scene context, semantic relationships, gaze direction, and object dissimilarity. However, none of these models consider the temporal nature of gaze shifts during image observation. We introduce a novel saliency prediction model that learns to output saliency maps in sequential time intervals by exploiting human temporal attention patterns. Our approach locally modulates the saliency predictions by combining the learned temporal maps. Our experiments show that our method outperforms the state-of-the-art models, including a multi-duration saliency model, on the SALICON benchmark. Our code will be publicly available on GitHub.
title TempSAL -- Uncovering Temporal Information for Deep Saliency Prediction
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2301.02315