Fixation prediction for advertising images: Dataset and benchmark

Song, Liang; Liu, Ruihang; Qian, Jiansheng

doi:10.1016/j.jvcir.2021.103356

Cited by 6 publications

(6 citation statements)

References 57 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, the saliency prediction algorithm trained on natural images has a strong ability to extract pattern information, but it is not sensitive to text and brand information, and the saliency prediction performance will be reduced when applied to advertising images. Method [20] shows that text enhancement based on the saliency prediction algorithm of natural images can significantly improve the performance of the algorithm on advertising images. Therefore, on the basis of the lightweight natural image saliency prediction method, i.e.…”

Section: Proposed Methodsmentioning

confidence: 99%

“…Humans have a strong centrality when viewing natural images. Research in advertising image saliency dataset ADD1000 [20] proves that human gaze on advertising images also has obvious center bias. In machine vision, this imaging center is called the extended focus or perspective projection center.…”

Section: Proposed Methodsmentioning

confidence: 99%

“…Since the purpose of this paper is to finally get the saliency region, it is not necessary to accurately box the text region, so the text candidate region is taken as the text saliency confidence score. A combination of traditional saliency and text region detection algorithm is used to extract text saliency, and paper [20] has proved that this mode can improve the accuracy of text detection.…”

Section: A Text Saliency Confidence Score Extraction Based On Image I...mentioning

confidence: 99%

“…The saliency algorithm in this paper is aimed at advertising images, so ADD1000 [20], the saliency dataset of advertising images, is used in this paper. This dataset contains 1000 advertising images, which are divided into four dimensions: (calm type, exciting type), (retro type, fashion type), (artistic type, practical type), and (emotional type, rational type), with a total of eight attributes.…”

Section: Experimental Settings a Datasetsmentioning

confidence: 99%

See 3 more Smart Citations

Advertising Image Saliency Prediction Method Based on Score Level Fusion

Kou

Liu

et al. 2023

IEEE Access

Self Cite

View full text Add to dashboard Cite

At present, visual saliency prediction algorithms have been developed more and more mature, but most of the current saliency prediction algorithms are aimed at natural images. Due to the inconsistency of elements and features between natural images and advertising images, the existing saliency prediction algorithms show poor robustness and low inference speed to advertising images, which severely limits its commercial application in advertising design and evaluation. In view of this, a saliency prediction algorithm for advertisement images is proposed in this paper. In the feature extraction stage, two text candidate regions based on intensity feature and improved MESR algorithm are first obtained and further integrated to produce a two-dimensional text confidence score. Meanwhile, a saliency confidence score is also obtained by an improved natural image saliency prediction network. Then, the score level fusion strategy was adopted to fuse the two confidence scores to get the final saliency prediction map. The experimental results show that the proposed model has good accuracy and robustness in advertising images, as well as the most remarkable inference speed, which can meet the demand for real-time performance of advertising image saliency prediction, leading to great practical and commercial value.INDEX TERMS Saliency prediction, eye-gaze assessment, advertising image, score level fusion.

show abstract

Section: Proposed Methodsmentioning

confidence: 99%

Section: Proposed Methodsmentioning

confidence: 99%

Section: A Text Saliency Confidence Score Extraction Based On Image I...mentioning

confidence: 99%

Section: Experimental Settings a Datasetsmentioning

confidence: 99%

See 2 more Smart Citations

Advertising Image Saliency Prediction Method Based on Score Level Fusion

Kou

Liu

et al. 2023

IEEE Access

Self Cite

View full text Add to dashboard Cite

show abstract

“…A gaze-points prediction method for advertising images was proposed. The method includes a CNNs-based model for saliency prediction of the multi-text advertising images [33].…”

Section: Strengths Weaknessesmentioning

confidence: 99%

Significant Measures of Gaze and Pupil Movement for Evaluating Empathy between Viewers and Digital Content

Zhang

Park

Cho

et al. 2022

Sensors

View full text Add to dashboard Cite

The success of digital content depends largely on whether viewers empathize with stories and narratives. Researchers have investigated the elements that may elicit empathy from viewers. Empathic response involves affective and cognitive processes and is expressed through multiple verbal and nonverbal modalities. Specifically, eye movements communicate emotions and intentions and may reflect an empathic status. This study explores feature changes in eye movements when a viewer empathizes with the video’s content. Seven feature variables of eye movements (change of pupil diameter, peak pupil dilation, very short, mid, over long fixation duration, saccadic amplitude, and saccadic count) were extracted from 47 participants who viewed eight videos (four empathic videos and four non-empathic videos) distributed in a two-dimensional emotion axis (arousal and valence). The results showed that viewers’ saccadic amplitude and peak pupil dilation in the eigenvalues of eye movements increased in the empathic condition. The fixation time and pupil size change showed limited significance, and whether there were asymmetric pupil responses between the left and right pupils remained inconclusive. Our investigation suggests that saccadic amplitude and peak pupil dilation are reliable measures for recognizing whether viewers empathize with content. The findings provide physiological evidence based on eye movements that both affective and cognitive processes accompany empathy during media consumption.

show abstract