{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,3]],"date-time":"2026-06-03T07:32:04Z","timestamp":1780471924671,"version":"3.54.1"},"reference-count":40,"publisher":"Springer Science and Business Media LLC","issue":"4","license":[{"start":{"date-parts":[[2023,12,19]],"date-time":"2023-12-19T00:00:00Z","timestamp":1702944000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2023,12,19]],"date-time":"2023-12-19T00:00:00Z","timestamp":1702944000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100003252","name":"Lund University","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100003252","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Behav Res"],"abstract":"<jats:title>Abstract<\/jats:title><jats:p>We present a deep learning method for accurately localizing the center of a single corneal reflection (CR) in an eye image. Unlike previous approaches, we use a convolutional neural network (CNN) that was trained solely using synthetic data. Using only synthetic data has the benefit of completely sidestepping the time-consuming process of manual annotation that is required for supervised training on real eye images. To systematically evaluate the accuracy of our method, we first tested it on images with synthetic CRs placed on different backgrounds and embedded in varying levels of noise. Second, we tested the method on two datasets consisting of high-quality videos captured from real eyes. Our method outperformed state-of-the-art algorithmic methods on real eye images with a 3\u201341.5% reduction in terms of spatial precision across data sets, and performed on par with state-of-the-art on synthetic images in terms of spatial accuracy. We conclude that our method provides a precise method for CR center localization and provides a solution to the data availability problem, which is one of the important common roadblocks in the development of deep learning models for gaze estimation. Due to the superior CR center localization and ease of application, our method has the potential to improve the accuracy and precision of CR-based eye trackers.<\/jats:p>","DOI":"10.3758\/s13428-023-02297-w","type":"journal-article","created":{"date-parts":[[2023,12,19]],"date-time":"2023-12-19T22:01:36Z","timestamp":1703023296000},"page":"3226-3241","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":9,"title":["Precise localization of corneal reflections in eye images using deep learning trained on synthetic data"],"prefix":"10.3758","volume":"56","author":[{"given":"Sean Anthony","family":"Byrne","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Marcus","family":"Nystr\u00f6m","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Virmarie","family":"Maquiling","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Enkelejda","family":"Kasneci","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4672-8756","authenticated-orcid":false,"given":"Diederick C.","family":"Niehorster","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2023,12,19]]},"reference":[{"issue":"6","key":"2297_CR1","doi-asserted-by":"publisher","first-page":"2480","DOI":"10.3758\/s13428-018-1026-7","volume":"50","author":"AD Barsingerhorn","year":"2018","unstructured":"Barsingerhorn, A. D., Boonstra, F. N., & Goossens, J. (2018). Development and validation of a high speed stereoscopic eyetracker. Behavior Research Methods, 50(6), 2480\u20132497. https:\/\/doi.org\/10.3758\/s13428-018-1026-7","journal-title":"Behavior Research Methods"},{"key":"2297_CR2","doi-asserted-by":"crossref","unstructured":"Cerrolaza, J. J., Villanueva, A., Villanueva, M., & Cabeza, R. (2012). Error characterization and compensation in eye tracking systems. Proceedings of the symposium on eye tracking research and applications, 205\u2013208","DOI":"10.1145\/2168556.2168595"},{"key":"2297_CR3","unstructured":"Cheng, Y.,Wang, H., Bao, Y., & Lu, F. (2021). Appearance-based gaze estimation with deep learning: A review and benchmark. CoRR, abs\/2104.12668. arXiv:2104.12668"},{"key":"2297_CR4","doi-asserted-by":"crossref","unstructured":"Chugh, S., Brousseau, B., Rose, J., & Eizenman, M. (2021). Detection and correspondence matching of corneal reflections for eye tracking using deep learning. 2020 25th International Conference on Pattern Recognition (ICPR), 2210\u2013 2217","DOI":"10.1109\/ICPR48806.2021.9412066"},{"key":"2297_CR5","unstructured":"Fuhl, W., Santini, T., Kasneci, G., Rosenstiel, W., & Kasneci, E. (2017). Pupilnet v2.0: Convolutional neural networks for cpu based real time robust pupil detection"},{"issue":"4","key":"2297_CR6","doi-asserted-by":"publisher","first-page":"506","DOI":"10.1364\/OPTICA.6.000506","volume":"6","author":"S Helgadottir","year":"2019","unstructured":"Helgadottir, S., Argun, A., & Volpe, G. (2019). Digital video microscopy enhanced by deep learning. Optica, 6(4), 506\u2013513.","journal-title":"Optica"},{"key":"2297_CR7","unstructured":"Holmqvist, K., Nystr\u00f6m, N., Andersson, R., Dewhurst, R., Jarodzka, H., & van de Weijer, J. (2011). Eye tracking: A comprehensive guide to methods and measures. Oxford University Press"},{"issue":"5","key":"2297_CR8","doi-asserted-by":"publisher","first-page":"2098","DOI":"10.3758\/s13428-020-01363-x","volume":"52","author":"K Holmqvist","year":"2020","unstructured":"Holmqvist, K., & Blignaut, P. (2020). Small eye movements cannot be reliably measured by video based p-cr eye-trackers. Behavior research methods, 52(5), 2098\u20132121.","journal-title":"Behavior research methods"},{"key":"2297_CR9","doi-asserted-by":"crossref","unstructured":"Holmqvist, K., Nystr\u00f6m, M., & Mulvey, F. (2012). Eye tracker data quality: What it is and how to measure it. Proceedings of the Symposium on Eye Tracking Research and Applications, 45\u2013 52","DOI":"10.1145\/2168556.2168563"},{"key":"2297_CR10","doi-asserted-by":"crossref","unstructured":"Hooge, I. T. C., Niehorster, D. C., Hessels, R. S., Benjamins, J. S., & Nystr\u00f6m, M. (2022). How robust are wearable eye trackers to slow and fast head and body movements? Behavior Research Methods, 1\u201315","DOI":"10.3758\/s13428-022-02010-3"},{"issue":"5","key":"2297_CR11","doi-asserted-by":"publisher","first-page":"1986","DOI":"10.3758\/s13428-020-01512-2","volume":"53","author":"ITC Hooge","year":"2021","unstructured":"Hooge, I. T. C., Niehorster, D. C., Hessels, R. S., Cleveland, D., & Nystr\u00f6m, M. (2021). The pupil-size artefact (psa) across time, viewing direction, and different eye trackers. Behavior Research Methods, 53(5), 1986\u20132006.","journal-title":"Behavior Research Methods"},{"issue":"5","key":"2297_CR12","doi-asserted-by":"publisher","first-page":"1864","DOI":"10.3758\/s13428-017-0955-x","volume":"50","author":"ITC Hooge","year":"2018","unstructured":"Hooge, I. T. C., Niehorster, D. C., Nystr\u00f6m, M., Andersson, R., & Hessels, R. S. (2018). Is human classification by experienced untrained observers a gold standard in fixation detection? Behavior Research Methods, 50(5), 1864\u20131881.","journal-title":"Behavior Research Methods"},{"issue":"3","key":"2297_CR13","doi-asserted-by":"publisher","first-page":"1387","DOI":"10.3758\/s13428-019-01305-2","volume":"52","author":"B Hosp","year":"2020","unstructured":"Hosp, B., Eivazi, S., Maurer, M., Fuhl, W., Geisler, D., & Kasneci, E. (2020). Remoteeye: An open-source high-speed remote eye tracker. Behavior Research Methods, 52(3), 1387\u20131401. https:\/\/doi.org\/10.3758\/s13428-019-01305-2","journal-title":"Behavior Research Methods"},{"key":"2297_CR14","doi-asserted-by":"publisher","unstructured":"Ivanchenko, D., Rifai, K., Hafed, Z. M., & Schaeffel, F. (2021). A low-cost, high-performance video-based binocular eye tracker for psychophysical research. Journal of Eye Movement Research, 14 (3). https:\/\/doi.org\/10.16910\/jemr.14.3.3","DOI":"10.16910\/jemr.14.3.3"},{"key":"2297_CR15","unstructured":"Kingma, D. P., & Ba, J. (2017). Adam: A method for stochastic optimization"},{"key":"2297_CR16","doi-asserted-by":"publisher","unstructured":"Li, D., Winfield, D., & Parkhurst, D. (2005). Starburst: A hybrid algorithm for video-based eye tracking combining feature-based and model-based approaches. 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR\u201905) - Workshops, 79\u201379. https:\/\/doi.org\/10.1109\/CVPR.2005.531","DOI":"10.1109\/CVPR.2005.531"},{"key":"2297_CR17","doi-asserted-by":"publisher","first-page":"309","DOI":"10.1109\/TBME.1974.324318","volume":"4","author":"J Merchant","year":"1974","unstructured":"Merchant, J., Morrissette, R., & Porterfield, J. L. (1974). Remote measurement of eye direction allowing subject motion over one cubic foot of space. IEEE Transactions On Biomedical Engineering, 4, 309\u2013317.","journal-title":"IEEE Transactions On Biomedical Engineering"},{"issue":"1","key":"2297_CR18","doi-asserted-by":"publisher","first-page":"011310","DOI":"10.1063\/5.0034891","volume":"8","author":"B Midtvedt","year":"2021","unstructured":"Midtvedt, B., Helgadottir, S., Argun, A., Pineda, J., Midtvedt, D., & Volpe, G. (2021). Quantitative digital microscopy with deep learning. Applied Physics Reviews, 8(1), 011310.","journal-title":"Applied Physics Reviews"},{"key":"2297_CR19","doi-asserted-by":"crossref","unstructured":"Midtvedt, B., Pineda, J., Sk\u00e4rberg, F., Ols\u00e9n, E., Bachimanchi, H., Wes\u00e9n, E., Midtvedt, & D. (2022). Single-shot self-supervised object detection in micro-scopy. Nature Communications, 13(1), 1\u201313.","DOI":"10.1038\/s41467-022-35004-y"},{"key":"2297_CR20","doi-asserted-by":"publisher","first-page":"54","DOI":"10.3758\/BF03200567","volume":"29","author":"JB Mulligan","year":"1997","unstructured":"Mulligan, J. B. (1997). Image processing for improved eye-tracking accuracy. Behavior Research Methods, Instruments, & Computers, 29, 54\u201365. https:\/\/doi.org\/10.3758\/BF03200567","journal-title":"Behavior Research Methods, Instruments, & Computers"},{"issue":"3","key":"2297_CR21","doi-asserted-by":"publisher","first-page":"1244","DOI":"10.3758\/s13428-019-01314-1","volume":"52","author":"DC Niehorster","year":"2020","unstructured":"Niehorster, D. C., Hessels, R. S., & Benjamins, J. S. (2020). Glassesviewer: Open-source software for viewing and analyzing data from the tobii pro glasses 2 eye tracker. Behavior Research Methods, 52(3), 1244\u20131253.","journal-title":"Behavior Research Methods"},{"key":"2297_CR22","doi-asserted-by":"publisher","unstructured":"Niehorster, D. C., & Nystr\u00f6m, M. (2018). Microsaccade detection using pupil and corneal reflection signals. Proceedings of the 2018 ACM Symposium on Eye Tracking Research & Applications. https:\/\/doi.org\/10.1145\/3204493.3204573","DOI":"10.1145\/3204493.3204573"},{"issue":"3","key":"2297_CR23","doi-asserted-by":"publisher","first-page":"1140","DOI":"10.3758\/s13428-019-01307-0","volume":"52","author":"D. C Niehorster","year":"2020","unstructured":"Niehorster, D. . C., Santini, T., Hessels, R. . S., Hooge, I. .T. .C., Kasneci, E., & Nystr\u00f6m, M. (2020a). The impact of slippage on the data quality of head-worn eye trackers. Behavior Research Methods, 52(3), 1140\u20131160. https:\/\/doi.org\/10.3758\/s13428-019-01307-0","journal-title":"Behavior Research Methods"},{"issue":"6","key":"2297_CR24","doi-asserted-by":"publisher","first-page":"2515","DOI":"10.3758\/s13428-020-01400-9","volume":"52","author":"DC Niehorster","year":"2020","unstructured":"Niehorster, D. C., Zemblys, R., Beelders, T., & Holmqvist, K. (2020b). Characterizing gaze position signals and synthesizing noise during fixations in eye-tracking data. Behavior Research Methods, 52(6), 2515\u20132534. https:\/\/doi.org\/10.3758\/s13428-020-01400-9","journal-title":"Behavior Research Methods"},{"issue":"1","key":"2297_CR25","doi-asserted-by":"publisher","first-page":"311","DOI":"10.3758\/s13428-020-01414-3","volume":"53","author":"DC Niehorster","year":"2021","unstructured":"Niehorster, D. C., Zemblys, R., & Holmqvist, K. (2021). Is apparent fixational drift in eye-tracking data due to filters or eyeball rotation? Behavior Research Methods, 53(1), 311\u2013324. https:\/\/doi.org\/10.3758\/s13428-020-01414-3","journal-title":"Behavior Research Methods"},{"key":"2297_CR26","doi-asserted-by":"publisher","unstructured":"Niu, L., Gu, Z., Ye, J., & Dong, Q. (2022). Real-time localization and matching of corneal reflections for eye gaze estimation via a lightweight network. The Ninth International Symposium of Chinese CHI, 33\u201340. https:\/\/doi.org\/10.1145\/3490355.3490359","DOI":"10.1145\/3490355.3490359"},{"key":"2297_CR27","doi-asserted-by":"crossref","unstructured":"Nystr\u00f6m, M., Niehorster, D.C., Andersson, R., Hessels, R.S., & Hooge, I.T.C. (2022). The amplitude of small eye movements can be accurately estimated with video-based eye trackers. Behavior Research Methods, 1\u201313","DOI":"10.3758\/s13428-021-01780-6"},{"key":"2297_CR28","doi-asserted-by":"crossref","unstructured":"Parthasarathy, R. (2012). Rapid, accurate particle tracking by calculation of radial symmetry centers. Nature methods, 9(7), 724\u2013 726.","DOI":"10.1038\/nmeth.2071"},{"key":"2297_CR29","unstructured":"Per\u00e9z, A., C\u00f3rdoba, M. L., Garcia, A., M\u00e9ndez, R., Munoz,M., Pedraza, J. L., & Sanchez, F. (2003). A precise eye-gaze detection and tracking system. Proceedings of The 11th International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision, 1-4. http:\/\/wscg.zcu.cz\/wscg2003\/Papers2003\/A83.pdf"},{"key":"2297_CR30","doi-asserted-by":"publisher","unstructured":"Pineda, J., Midtvedt, B., Bachimanchi, H., N\u00f3e, S., Midtvedt, D., Volpe, G., & Manzo, C. (2022). Geometric deep learning reveals the spatiotemporal fingerprint of microscopic motion. https:\/\/doi.org\/10.48550\/ARXIV.2202.06355","DOI":"10.48550\/ARXIV.2202.06355"},{"key":"2297_CR31","doi-asserted-by":"publisher","unstructured":"San Agustin, J., Skovsgaard, H., Mollenbach, E., Barret, M., Tall, M., Hansen, D. W., & Hansen, J. P. (2010). Evaluation of a low-cost open source gaze tracker. Proceedings of the 2010 Symposium on Eye-Tracking Research & Applications, 77\u201380. https:\/\/doi.org\/10.1145\/1743666.1743685","DOI":"10.1145\/1743666.1743685"},{"key":"2297_CR32","doi-asserted-by":"publisher","first-page":"239","DOI":"10.1117\/12.189136","volume":"2350","author":"MR Shortis","year":"1994","unstructured":"Shortis, M. R., Clarke, T. A., & Short, T. (1994). Comparison of some techniques for the subpixel location of discrete target images. Videometrics III, 2350, 239\u2013250.","journal-title":"Videometrics III"},{"key":"2297_CR33","doi-asserted-by":"crossref","unstructured":"Shrivastava, A., Pfister, T., Tuzel, O., Susskind, J., Wang, W., & Webb, R. (2016). Learning from simulated and unsupervised images through adversarial training. CoRR abs\/1612.07828. arXiv:1612.07828","DOI":"10.1109\/CVPR.2017.241"},{"issue":"2","key":"2297_CR34","doi-asserted-by":"publisher","first-page":"137","DOI":"10.3758\/BF03204486","volume":"25","author":"DM Stampe","year":"1993","unstructured":"Stampe, D. M. (1993). Heuristic filtering and reliable calibration methods for video-based pupil-tracking systems. Behavior Research Methods, Instruments, & Computers, 25(2), 137\u2013142.","journal-title":"Behavior Research Methods, Instruments, & Computers"},{"key":"2297_CR35","doi-asserted-by":"publisher","unstructured":"Sugano, Y., Matsushita, Y., & Sato, Y. (2014). Learning-by-synthesis for appearance-based 3d gaze estimation. 2014 IEEE Conference on Computer Vision and Pattern Recognition, 1821\u20131828. https:\/\/doi.org\/10.1109\/CVPR.2014.235","DOI":"10.1109\/CVPR.2014.235"},{"key":"2297_CR36","doi-asserted-by":"crossref","unstructured":"Wood, E., Baltrusaitis, T., Zhang, X., Sugano, Y., Robinson, P., & Bulling, A. (2015). Rendering of eyes for eye-shape registration and gaze estimation. CoRR, abs\/1505.05916. arXiv:1505.05916","DOI":"10.1109\/ICCV.2015.428"},{"key":"2297_CR37","doi-asserted-by":"publisher","unstructured":"Wood, E., Baltru\u0161aitis, T., Morency, L.-P., Robinson, P., & Bulling, A. (2016). Learning an appearance-based gaze estimator from one million synthesised images. Proceedings of the Ninth Biennial ACM Symposium on Eye Tracking Research & Applications, 131\u2013138. https:\/\/doi.org\/10.1145\/2857491.2857492","DOI":"10.1145\/2857491.2857492"},{"key":"2297_CR38","doi-asserted-by":"publisher","unstructured":"Wu, R.-J., Clark, A., Cox, M., Intoy, J., Jolly, P., Zhao, Z., & Rucci, M. (2022). High-resolution eye-tracking via digital imaging of purkinje reflections. bioRxiv. https:\/\/doi.org\/10.1101\/2022.08.16.504076","DOI":"10.1101\/2022.08.16.504076"},{"key":"2297_CR39","doi-asserted-by":"crossref","unstructured":"Wu, Z., Rajendran, S., Van As, T., Badrinarayanan, V., & Rabinovich, A. (2019). Eyenet: A multi-task deep network for off-axis eye gaze estimation. 2019 IEEE\/CVF International Conference on Computer Vision Workshop (ICCVW), 3683\u20133687","DOI":"10.1109\/ICCVW.2019.00455"},{"key":"2297_CR40","doi-asserted-by":"publisher","unstructured":"Zimmermann, J., Vazquez, Y., Glimcher, P. W., Pesaran, B., & Louie, K. (2016). Oculomatic: High speed, reliable, and accurate open-source eye tracking for humans and non-human primates. Journal of Neuroscience Methods, 270, 138\u2013146. https:\/\/doi.org\/10.1016\/j.jneumeth.2016.06.016","DOI":"10.1016\/j.jneumeth.2016.06.016"}],"container-title":["Behavior Research Methods"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.3758\/s13428-023-02297-w.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.3758\/s13428-023-02297-w\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.3758\/s13428-023-02297-w.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,5,30]],"date-time":"2024-05-30T14:14:19Z","timestamp":1717078459000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.3758\/s13428-023-02297-w"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,12,19]]},"references-count":40,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2024,6]]}},"alternative-id":["2297"],"URL":"https:\/\/doi.org\/10.3758\/s13428-023-02297-w","relation":{},"ISSN":["1554-3528"],"issn-type":[{"value":"1554-3528","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,12,19]]},"assertion":[{"value":"15 November 2023","order":1,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"19 December 2023","order":2,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}}]}}