arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1810.04456cs.CV
收藏

场景自由观察中显著性模型对人类注视的不变性分析

Invariance Analysis of Saliency Models versus Human Gaze During Scene Free Viewing

Shanghai Jiao Tong University(上海交通大学);University of Central Florida(中佛罗里达大学)

Zhaohui Che, Ali Borji, Guangtao Zhai, Xiongkuo Min

摘要:

目前,大多数关于人类注视和显著性建模的研究都使用高质量的刺激。然而,在现实世界中,捕获的图像在整个获取、传输和显示链期间经历各种类型的失真。某些失真类型包括运动模糊、照明变化和旋转。尽管很少的努力,无处不在的扭曲视觉注意和显着性模型的影响还没有系统地研究。在本文中,我们首先创建了一个大规模的数据库,其中包括10个观察者的眼睛运动超过1900图像退化的19种类型的失真。其次,通过分析眼动和显着性模型,我们发现:a)观察者在失真与原始图像上看不同的位置,b)显着性模型的性能在失真图像上受到严重阻碍,最大的性能下降属于旋转和剪切失真。最后,我们调查的有效性,不同的扭曲时,作为数据增强转换。实验结果表明,一些有效的数据增强变换能够保持参考图像的人眼视线,提高深度显著性模型的抗失真能力,而一些无效的变换会严重改变人眼视线,降低模型的性能。

英文摘要:

Most of current studies on human gaze and saliency modeling have used high-quality stimuli. In real world, however, captured images undergo various types of distortions during the whole acquisition, transmission, and displaying chain. Some distortion types include motion blur, lighting variations and rotation. Despite few efforts, influences of ubiquitous distortions on visual attention and saliency models have not been systematically investigated. In this paper, we first create a large-scale database including eye movements of 10 observers over 1900 images degraded by 19 types of distortions. Second, by analyzing eye movements and saliency models, we find that: a) observers look at different locations over distorted versus original images, and b) performances of saliency models are drastically hindered over distorted images, with the maximum performance drop belonging to Rotation and Shearing distortions. Finally, we investigate the effectiveness of different distortions when serving as data augmentation transformations. Experimental results verify that some useful data augmentation transformations which preserve human gaze of reference images can improve deep saliency models against distortions, while some invalid transformations which severely change human gaze will degrade the performance.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑