V-ir-Net : A Novel Neural Network for Pupil and Corneal Reflection Detection trained on Simulated Light Distributions
(2023) p.1-7- Abstract
- Deep learning has shown promise for gaze estimation in Virtual Reality (VR) and other head-mounted applications, but such models are hard to train due to lack of available data. Here we introduce a novel method to train neural networks for gaze estimation using synthetic images that model the light distributions captured in a P-CR setup. We tested our model on a dataset of real eye images from a VR setup, achieving 76% accuracy which is close to the state-of-the-art model which was trained on the dataset itself. The localization error for CRs was 1.56 pixels and 2.02 pixels for the pupil, which is on par with state-of-the-art. Our approach allowed inference on the whole dataset without sacrificing data for model training. Our method... (More)
- Deep learning has shown promise for gaze estimation in Virtual Reality (VR) and other head-mounted applications, but such models are hard to train due to lack of available data. Here we introduce a novel method to train neural networks for gaze estimation using synthetic images that model the light distributions captured in a P-CR setup. We tested our model on a dataset of real eye images from a VR setup, achieving 76% accuracy which is close to the state-of-the-art model which was trained on the dataset itself. The localization error for CRs was 1.56 pixels and 2.02 pixels for the pupil, which is on par with state-of-the-art. Our approach allowed inference on the whole dataset without sacrificing data for model training. Our method provides a cost-efficient and lightweight training alternative, eliminating the need for hand-labeled data. It offers flexible customization, e.g. adapting to different illuminator configurations, with minimal code changes. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/eb3a8c08-a6c1-45b3-8c23-32afd9f23de0
- author
- Maquiling, Virmarie ; Byrne, Sean Anthony ; Nyström, Marcus LU ; Kasneci, Enkelejda and Niehorster, Diederick C. LU
- organization
- publishing date
- 2023-09-26
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- host publication
- MobileHCI '23 Companion : Proceedings of the 25th International Conference on Mobile Human-Computer Interaction - Proceedings of the 25th International Conference on Mobile Human-Computer Interaction
- editor
- Komninos, Andreas ; Santoro, Carmen ; Gavalas, Damianos ; Schoening, Johannes ; Matera, Maristella and Leiva, Luis A.
- article number
- 23
- pages
- 7 pages
- publisher
- Association for Computing Machinery (ACM)
- external identifiers
-
- scopus:85174318528
- ISBN
- 978-1-4503-9924-1
- DOI
- 10.1145/3565066.3608690
- language
- English
- LU publication?
- yes
- id
- eb3a8c08-a6c1-45b3-8c23-32afd9f23de0
- alternative location
- https://dl.acm.org/doi/10.1145/3565066.3608690
- date added to LUP
- 2023-09-27 10:33:59
- date last changed
- 2023-12-11 15:19:46
@inproceedings{eb3a8c08-a6c1-45b3-8c23-32afd9f23de0, abstract = {{Deep learning has shown promise for gaze estimation in Virtual Reality (VR) and other head-mounted applications, but such models are hard to train due to lack of available data. Here we introduce a novel method to train neural networks for gaze estimation using synthetic images that model the light distributions captured in a P-CR setup. We tested our model on a dataset of real eye images from a VR setup, achieving 76% accuracy which is close to the state-of-the-art model which was trained on the dataset itself. The localization error for CRs was 1.56 pixels and 2.02 pixels for the pupil, which is on par with state-of-the-art. Our approach allowed inference on the whole dataset without sacrificing data for model training. Our method provides a cost-efficient and lightweight training alternative, eliminating the need for hand-labeled data. It offers flexible customization, e.g. adapting to different illuminator configurations, with minimal code changes.}}, author = {{Maquiling, Virmarie and Byrne, Sean Anthony and Nyström, Marcus and Kasneci, Enkelejda and Niehorster, Diederick C.}}, booktitle = {{MobileHCI '23 Companion : Proceedings of the 25th International Conference on Mobile Human-Computer Interaction}}, editor = {{Komninos, Andreas and Santoro, Carmen and Gavalas, Damianos and Schoening, Johannes and Matera, Maristella and Leiva, Luis A.}}, isbn = {{978-1-4503-9924-1}}, language = {{eng}}, month = {{09}}, pages = {{1--7}}, publisher = {{Association for Computing Machinery (ACM)}}, title = {{V-ir-Net : A Novel Neural Network for Pupil and Corneal Reflection Detection trained on Simulated Light Distributions}}, url = {{http://dx.doi.org/10.1145/3565066.3608690}}, doi = {{10.1145/3565066.3608690}}, year = {{2023}}, }