Understanding Image Impressiveness Inspired by Instantaneous Human Perceptual Cues

Jufeng Yang, Yan Sun, Jie Liang, Yongliang Yang, Ming-Ming Cheng

Research output: Contribution to conferencePaper

Abstract

With the explosion of visual information nowadays, millions of digital images are available to the users. How to efficiently explore a large set of images and retrieve useful information thus becomes extremely important. Unfortunately only some of the images can impress the user at first glance. Others that make little sense in human perception are often discarded, while still costing valuable time and space. Therefore, it is significant to identify these two kinds of images for relieving the load of online repositories and accelerating information retrieval process. However, most of the existing imageproperties,e.g.,memorabilityandpopularity,arebased on repeated human interactions, which limit the research and application of evaluating image quality in terms of instantaneous impression. In this paper, we propose a novel image property, called impressiveness, that measures how images impress people witha short-term contact. This isbased on an impression-driven model inspired by a number of important human perceptual cues. To achieve this, we first collect three datasets in various domains, which are labeled according to theinstantaneoussensationoftheannotators.Thenweinvestigate the impressiveness property via six established human perceptual cues as well as the corresponding features from pixel to semantic levels. Sequentially, we verify the consistencyoftheimpressiveness whichcanbequantitativelymeasured by multiple visual representations, and evaluate their latent relationships. Finally, we apply the proposed impressiveness property to rank the images for an efficient image recommendation system.
LanguageEnglish
Number of pages8
StatusAccepted/In press - 2018
EventAAAI Conference on Artificial Intelligence 2018 -
Duration: 2 Feb 20187 Feb 2018

Conference

ConferenceAAAI Conference on Artificial Intelligence 2018
Period2/02/187/02/18

Fingerprint

Image understanding
Recommender systems
Information retrieval
Image quality
Explosions
Pixels
Semantics

Cite this

Yang, J., Sun, Y., Liang, J., Yang, Y., & Cheng, M-M. (Accepted/In press). Understanding Image Impressiveness Inspired by Instantaneous Human Perceptual Cues. Paper presented at AAAI Conference on Artificial Intelligence 2018, .

Understanding Image Impressiveness Inspired by Instantaneous Human Perceptual Cues. / Yang, Jufeng; Sun, Yan; Liang, Jie; Yang, Yongliang; Cheng, Ming-Ming.

2018. Paper presented at AAAI Conference on Artificial Intelligence 2018, .

Research output: Contribution to conferencePaper

Yang, J, Sun, Y, Liang, J, Yang, Y & Cheng, M-M 2018, 'Understanding Image Impressiveness Inspired by Instantaneous Human Perceptual Cues' Paper presented at AAAI Conference on Artificial Intelligence 2018, 2/02/18 - 7/02/18, .
Yang J, Sun Y, Liang J, Yang Y, Cheng M-M. Understanding Image Impressiveness Inspired by Instantaneous Human Perceptual Cues. 2018. Paper presented at AAAI Conference on Artificial Intelligence 2018, .
Yang, Jufeng ; Sun, Yan ; Liang, Jie ; Yang, Yongliang ; Cheng, Ming-Ming. / Understanding Image Impressiveness Inspired by Instantaneous Human Perceptual Cues. Paper presented at AAAI Conference on Artificial Intelligence 2018, .8 p.
@conference{a3dbd8e3d8ab4ea88fba19fdc973c00d,
title = "Understanding Image Impressiveness Inspired by Instantaneous Human Perceptual Cues",
abstract = "With the explosion of visual information nowadays, millions of digital images are available to the users. How to efficiently explore a large set of images and retrieve useful information thus becomes extremely important. Unfortunately only some of the images can impress the user at first glance. Others that make little sense in human perception are often discarded, while still costing valuable time and space. Therefore, it is significant to identify these two kinds of images for relieving the load of online repositories and accelerating information retrieval process. However, most of the existing imageproperties,e.g.,memorabilityandpopularity,arebased on repeated human interactions, which limit the research and application of evaluating image quality in terms of instantaneous impression. In this paper, we propose a novel image property, called impressiveness, that measures how images impress people witha short-term contact. This isbased on an impression-driven model inspired by a number of important human perceptual cues. To achieve this, we first collect three datasets in various domains, which are labeled according to theinstantaneoussensationoftheannotators.Thenweinvestigate the impressiveness property via six established human perceptual cues as well as the corresponding features from pixel to semantic levels. Sequentially, we verify the consistencyoftheimpressiveness whichcanbequantitativelymeasured by multiple visual representations, and evaluate their latent relationships. Finally, we apply the proposed impressiveness property to rank the images for an efficient image recommendation system.",
author = "Jufeng Yang and Yan Sun and Jie Liang and Yongliang Yang and Ming-Ming Cheng",
year = "2018",
language = "English",
note = "AAAI Conference on Artificial Intelligence 2018 ; Conference date: 02-02-2018 Through 07-02-2018",

}

TY - CONF

T1 - Understanding Image Impressiveness Inspired by Instantaneous Human Perceptual Cues

AU - Yang, Jufeng

AU - Sun, Yan

AU - Liang, Jie

AU - Yang, Yongliang

AU - Cheng, Ming-Ming

PY - 2018

Y1 - 2018

N2 - With the explosion of visual information nowadays, millions of digital images are available to the users. How to efficiently explore a large set of images and retrieve useful information thus becomes extremely important. Unfortunately only some of the images can impress the user at first glance. Others that make little sense in human perception are often discarded, while still costing valuable time and space. Therefore, it is significant to identify these two kinds of images for relieving the load of online repositories and accelerating information retrieval process. However, most of the existing imageproperties,e.g.,memorabilityandpopularity,arebased on repeated human interactions, which limit the research and application of evaluating image quality in terms of instantaneous impression. In this paper, we propose a novel image property, called impressiveness, that measures how images impress people witha short-term contact. This isbased on an impression-driven model inspired by a number of important human perceptual cues. To achieve this, we first collect three datasets in various domains, which are labeled according to theinstantaneoussensationoftheannotators.Thenweinvestigate the impressiveness property via six established human perceptual cues as well as the corresponding features from pixel to semantic levels. Sequentially, we verify the consistencyoftheimpressiveness whichcanbequantitativelymeasured by multiple visual representations, and evaluate their latent relationships. Finally, we apply the proposed impressiveness property to rank the images for an efficient image recommendation system.

AB - With the explosion of visual information nowadays, millions of digital images are available to the users. How to efficiently explore a large set of images and retrieve useful information thus becomes extremely important. Unfortunately only some of the images can impress the user at first glance. Others that make little sense in human perception are often discarded, while still costing valuable time and space. Therefore, it is significant to identify these two kinds of images for relieving the load of online repositories and accelerating information retrieval process. However, most of the existing imageproperties,e.g.,memorabilityandpopularity,arebased on repeated human interactions, which limit the research and application of evaluating image quality in terms of instantaneous impression. In this paper, we propose a novel image property, called impressiveness, that measures how images impress people witha short-term contact. This isbased on an impression-driven model inspired by a number of important human perceptual cues. To achieve this, we first collect three datasets in various domains, which are labeled according to theinstantaneoussensationoftheannotators.Thenweinvestigate the impressiveness property via six established human perceptual cues as well as the corresponding features from pixel to semantic levels. Sequentially, we verify the consistencyoftheimpressiveness whichcanbequantitativelymeasured by multiple visual representations, and evaluate their latent relationships. Finally, we apply the proposed impressiveness property to rank the images for an efficient image recommendation system.

M3 - Paper

ER -