Selection of a best metric and evaluation of bottom-up visual saliency models
作者:
Highlights:
• Introduced a set of experiments to judge the biological plausibility of visual saliency models.
• Introduced a novel method to evaluate saliency map comparison metrics using a database of human fixation maps.
• Employed the introduced method to identify the best saliency map comparison metric.
• Examined nine well-known models of visual saliency using the best metric to identify the best visual saliency models.
摘要
•Introduced a set of experiments to judge the biological plausibility of visual saliency models.•Introduced a novel method to evaluate saliency map comparison metrics using a database of human fixation maps.•Employed the introduced method to identify the best saliency map comparison metric.•Examined nine well-known models of visual saliency using the best metric to identify the best visual saliency models.
论文关键词:Bottom-up saliency mechanism,Best comparison metric,Computer vision,Human visual system
论文评审过程:Received 7 August 2012, Revised 17 April 2013, Accepted 13 August 2013, Available online 29 August 2013.
论文官网地址:https://doi.org/10.1016/j.imavis.2013.08.004