Saccade Landing Point Prediction Based on Fine-Grained Learning Method
Entity
UAM. Departamento de Tecnología Electrónica y de las ComunicacionesPublisher
Institute of Electrical and Electronics Engineers Inc. (IEEE)Date
2021-04-01Citation
10.1109/ACCESS.2021.3070511
IEEE Access 9 (2021): 52474-52484
ISSN
2169-3536 (print)DOI
10.1109/ACCESS.2021.3070511Funded by
This work was supported in part by the Project BIBECA through MINECO/FEDER under Grant RTI2018-101248-B-100, in part by the Jose Castillejo Program through MINECO under Grant CAS17/00117, and in part by the National Institutes of Health (NIH) under Grant P30EY003790 and Grant R21EY023724.Project
Gobierno de España. Grant RTI2018-101248-B-100; Gobierno de España. Grant CAS17/00117Editor's Version
https://doi.org/10.1109/ACCESS.2021.3070511Subjects
Saccade; Fine-grained learning; LSTM; Recurrent neural networks; Gaze-contingent; TelecomunicacionesRights
© AuthorsAbstract
The landing point of a saccade defines the new fixation region, the new region of interest. We asked whether it was possible to predict the saccade landing point early in this very fast eye movement. This work proposes a new algorithm based on LSTM networks and a fine-grained loss function for saccade landing point prediction in real-world scenarios. Predicting the landing point is a critical milestone toward reducing the problems caused by display-update latency in gaze-contingent systems that make real-time changes in the display based on eye tracking. Saccadic eye movements are some of the fastest human neuro-motor activities with angular velocities of up to 1,000°/s. We present a comprehensive analysis of the performance of our method using a database with almost 220,000 saccades from 75 participants captured during natural viewing of videos. We include a comparison with state-of-the-art saccade landing point prediction algorithms. The results obtained using our proposed method outperformed existing approaches with improvements of up to 50% error reduction. Finally, we analyzed some factors that affected prediction errors including duration, length, age, and user intrinsic characteristics.
Files in this item
Google Scholar:Morales Moreno, Aythami
-
Costela, Francisco M.
-
Woods, Russell L.
This item appears in the following Collection(s)
Related items
Showing items related by title, author, creator and subject.