EM-Gaze: eye context correlation and metric learning for gaze estimation
Abstract In recent years, deep learning techniques have been used to estimate gaze—a significant task in computer vision and human-computer interaction. Previous studies have made significant achievements in predicting 2D or 3D gazes from monocular face images. This study presents a deep neural netw...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2023-05-01
|
| Series: | Visual Computing for Industry, Biomedicine, and Art |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s42492-023-00135-6 |
| _version_ | 1797832237449740288 |
|---|---|
| author | Jinchao Zhou Guoan Li Feng Shi Xiaoyan Guo Pengfei Wan Miao Wang |
| author_facet | Jinchao Zhou Guoan Li Feng Shi Xiaoyan Guo Pengfei Wan Miao Wang |
| author_sort | Jinchao Zhou |
| collection | DOAJ |
| description | Abstract In recent years, deep learning techniques have been used to estimate gaze—a significant task in computer vision and human-computer interaction. Previous studies have made significant achievements in predicting 2D or 3D gazes from monocular face images. This study presents a deep neural network for 2D gaze estimation on mobile devices. It achieves state-of-the-art 2D gaze point regression error, while significantly improving gaze classification error on quadrant divisions of the display. To this end, an efficient attention-based module that correlates and fuses the left and right eye contextual features is first proposed to improve gaze point regression performance. Subsequently, through a unified perspective for gaze estimation, metric learning for gaze classification on quadrant divisions is incorporated as additional supervision. Consequently, both gaze point regression and quadrant classification performances are improved. The experiments demonstrate that the proposed method outperforms existing gaze-estimation methods on the GazeCapture and MPIIFaceGaze datasets. |
| first_indexed | 2024-04-09T14:05:39Z |
| format | Article |
| id | doaj.art-9758ebc127ce40d18acdbf4183a9fef2 |
| institution | Directory Open Access Journal |
| issn | 2524-4442 |
| language | English |
| last_indexed | 2024-04-09T14:05:39Z |
| publishDate | 2023-05-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | Visual Computing for Industry, Biomedicine, and Art |
| spelling | doaj.art-9758ebc127ce40d18acdbf4183a9fef22023-05-07T11:04:31ZengSpringerOpenVisual Computing for Industry, Biomedicine, and Art2524-44422023-05-016111210.1186/s42492-023-00135-6EM-Gaze: eye context correlation and metric learning for gaze estimationJinchao Zhou0Guoan Li1Feng Shi2Xiaoyan Guo3Pengfei Wan4Miao Wang5State Key Laboratory of Virtual Reality Technology and Systems, Beihang UniversityState Key Laboratory of Virtual Reality Technology and Systems, Beihang UniversityKuaishou TechnologyKuaishou TechnologyKuaishou TechnologyState Key Laboratory of Virtual Reality Technology and Systems, Beihang UniversityAbstract In recent years, deep learning techniques have been used to estimate gaze—a significant task in computer vision and human-computer interaction. Previous studies have made significant achievements in predicting 2D or 3D gazes from monocular face images. This study presents a deep neural network for 2D gaze estimation on mobile devices. It achieves state-of-the-art 2D gaze point regression error, while significantly improving gaze classification error on quadrant divisions of the display. To this end, an efficient attention-based module that correlates and fuses the left and right eye contextual features is first proposed to improve gaze point regression performance. Subsequently, through a unified perspective for gaze estimation, metric learning for gaze classification on quadrant divisions is incorporated as additional supervision. Consequently, both gaze point regression and quadrant classification performances are improved. The experiments demonstrate that the proposed method outperforms existing gaze-estimation methods on the GazeCapture and MPIIFaceGaze datasets.https://doi.org/10.1186/s42492-023-00135-6Computer visionGaze estimationMetric learningAttentionMulti-task learning |
| spellingShingle | Jinchao Zhou Guoan Li Feng Shi Xiaoyan Guo Pengfei Wan Miao Wang EM-Gaze: eye context correlation and metric learning for gaze estimation Visual Computing for Industry, Biomedicine, and Art Computer vision Gaze estimation Metric learning Attention Multi-task learning |
| title | EM-Gaze: eye context correlation and metric learning for gaze estimation |
| title_full | EM-Gaze: eye context correlation and metric learning for gaze estimation |
| title_fullStr | EM-Gaze: eye context correlation and metric learning for gaze estimation |
| title_full_unstemmed | EM-Gaze: eye context correlation and metric learning for gaze estimation |
| title_short | EM-Gaze: eye context correlation and metric learning for gaze estimation |
| title_sort | em gaze eye context correlation and metric learning for gaze estimation |
| topic | Computer vision Gaze estimation Metric learning Attention Multi-task learning |
| url | https://doi.org/10.1186/s42492-023-00135-6 |
| work_keys_str_mv | AT jinchaozhou emgazeeyecontextcorrelationandmetriclearningforgazeestimation AT guoanli emgazeeyecontextcorrelationandmetriclearningforgazeestimation AT fengshi emgazeeyecontextcorrelationandmetriclearningforgazeestimation AT xiaoyanguo emgazeeyecontextcorrelationandmetriclearningforgazeestimation AT pengfeiwan emgazeeyecontextcorrelationandmetriclearningforgazeestimation AT miaowang emgazeeyecontextcorrelationandmetriclearningforgazeestimation |