首页> 外文期刊>Journal of Computers >3D Visual Attention for Stereoscopic Image Quality Assessment
【24h】

3D Visual Attention for Stereoscopic Image Quality Assessment

机译:立体图像质量评估的3D视觉注意

获取原文
获取外文期刊封面目录资料

摘要

Depth perception is one of the most importantcharacteristic in three-dimensional (3D) images differentfrom traditional two-dimensional (2D) images. Therefore,3D visual attention will be advantageous to improve 3Dvisual experience and particularly depth perception. In thispaper, we propose a 3D visual attention model forstereoscopic image quality assessment task. The proposedmodel is constructed based on 2D saliency model, centerbias, depth cue (foreground cue and background). Differentcombination and modulation means of the 3D visualattention model for quality assessment are investigated. Theexperimental results show that compared with otherschemes, the proposed 3D visual attention-based poolingscheme can achieve higher consistency with the subjectiveassessment of stereoscopic images.
机译:与传统的二维(2D)图像不同,深度感知是三维(3D)图像中最重要的特征之一。因此,3D视觉注意力将有利于改善3D视觉体验,尤其是深度感知。在本文中,我们提出了用于立体图像质量评估任务的3D视觉注意模型。该模型是基于二维显着性模型,中心偏差,深度提示(前景提示和背景)构建的。研究了用于质量评估的3D视觉注意模型的不同组合和调制方式。实验结果表明,与其他方案相比,所提出的基于视觉注意的3D视觉合并方案与立体图像的主观评估具有更高的一致性。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号