Scanpaths Generation for Target Search Based on Deep Learning

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

Scanpath is a sequence of gaze fixations changing with time when browsing something, which records eyes' movement dynamically. Accurate prediction of scanpaths can help computers better predict which area human pay attention to and drive the development of next-generation systems that can understand human behavior and needs. Therefore, this paper introduces an algorithm of scanpaths generation for target search. Based on one-shot learning network, the algorithm extracts the target map with task information, and then imports it into a visual model of the superior colliculus to predict the task-drive scanpaths. The results are similar to the gaze behavior of human eyes in shape, direction and length. At the same time, the algorithm also has the ability to predict the scanpaths based on unseen categories.

Original languageEnglish
Title of host publicationProceedings - 2020 Chinese Automation Congress, CAC 2020
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1443-1448
Number of pages6
ISBN (Electronic)9781728176871
DOIs
StatePublished - 6 Nov 2020
Event2020 Chinese Automation Congress, CAC 2020 - Shanghai, China
Duration: 6 Nov 20208 Nov 2020

Publication series

NameProceedings - 2020 Chinese Automation Congress, CAC 2020

Conference

Conference2020 Chinese Automation Congress, CAC 2020
Country/TerritoryChina
CityShanghai
Period6/11/208/11/20

Keywords

  • Computer Vision
  • Deep Learning
  • Scanpaths Generation
  • Target Search

Fingerprint

Dive into the research topics of 'Scanpaths Generation for Target Search Based on Deep Learning'. Together they form a unique fingerprint.

Cite this