Semi-automatic Annotation of Images Using Eye Gaze Data (SAIGA)
Semi-automatic Annotation of Images Using Eye Gaze Data (SAIGA)
复制标题
使用眼睛注视数据的图像半自动注释 (SAIGA)
DOI：
--
复制
发表时间：
2018
期刊：
影响因子：
--
通讯作者：
J. Amudha
中科院分区：
文献类型：
--
作者：
B. Gottimukkala;M. Praveen;P. Amruta;J. Amudha
视线跟踪是基于瞳孔运动的，是人机交互的有效媒介。该字段以多种方式使用，并且由于其增加的易用性和提高的准确性而越来越受欢迎。本文的主要目标是提出一个框架，将减轻图像注释的负担，使其更具交互性。图像通过物理描述其元数据进行注释。目前的系统给用户100%的自由，以标记的图像在他/她的自由裁量权，但非常繁琐和耗时。在这里，我们提出了半自动注释的图像使用眼睛凝视数据（SAIGA），一种方法，将有助于使用眼睛凝视数据来注释图像。SAIGA-建议的框架显示了如何花费在手动注释的时间和体力可以更好的一个大的价值。
Eye gaze tracking is based on pupil movement and is an effective medium for human–computer interaction. This field is utilized in several ways and is gaining popularity due to its increased ease of use and improved accuracy. The main objective of this paper is to present a framework that would assuage the burden of image annotations and make it more interactive. Images are annotated by physically describing their metadata. The current system gives the user 100% freedom to label the images at his/her discretion but is very tedious and time consuming. Here, we propose semi-automatic annotation of images using eye gaze data (SAIGA), an approach that would assist in using the eye gaze data to annotate images. SAIGA—the proposed framework shows how time and physical efforts spent on manual annotation can be bettered by a large value.