CI-Net: Appearance-Based Gaze Estimation via Cooperative Network
Facial occlusion and different appearances of both eyes in natural scenes can affect the accuracy of gaze estimation based on appearance. Therefore, this paper proposes a gaze estimation model based on cooperative network: CI-Net, including a consistency estimation network (C-Net) and inconsistency...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2022-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9841577/ |
| _version_ | 1811341987069034496 |
|---|---|
| author | Yuan Luo Jiangtao Chen Jian Chen |
| author_facet | Yuan Luo Jiangtao Chen Jian Chen |
| author_sort | Yuan Luo |
| collection | DOAJ |
| description | Facial occlusion and different appearances of both eyes in natural scenes can affect the accuracy of gaze estimation based on appearance. Therefore, this paper proposes a gaze estimation model based on cooperative network: CI-Net, including a consistency estimation network (C-Net) and inconsistency estimation network (I-Net). C-Net is used to estimate the Main gaze of the true gaze, and an attention mechanism is added to adaptively assign the weight between eyes and face features. The I-Net is used to estimate the Residual residuals based on true gaze. In addition, Cross attention module is designed in this paper, through which I-Net can selectively obtain information from C-Net, to obtain more accurate eyes directions. The experimental results in this paper show that the CI-Net gain lower angle errors than the current mainstream CNN methods under the condition of different appearance of both eyes and facial occlusion. |
| first_indexed | 2024-04-13T19:03:18Z |
| format | Article |
| id | doaj.art-320246339a9a4e98999ed67295216548 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-13T19:03:18Z |
| publishDate | 2022-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-320246339a9a4e98999ed672952165482022-12-22T02:34:02ZengIEEEIEEE Access2169-35362022-01-0110787397874610.1109/ACCESS.2022.31941239841577CI-Net: Appearance-Based Gaze Estimation via Cooperative NetworkYuan Luo0https://orcid.org/0000-0002-1405-817XJiangtao Chen1https://orcid.org/0000-0001-5048-3336Jian Chen2https://orcid.org/0000-0002-9512-4123Key Laboratory of Optoelectronic Information Sensing and Technology, Chongqing University of Posts and Telecommunications, Chongqing, ChianKey Laboratory of Optoelectronic Information Sensing and Technology, Chongqing University of Posts and Telecommunications, Chongqing, ChianKey Laboratory of Optoelectronic Information Sensing and Technology, Chongqing University of Posts and Telecommunications, Chongqing, ChianFacial occlusion and different appearances of both eyes in natural scenes can affect the accuracy of gaze estimation based on appearance. Therefore, this paper proposes a gaze estimation model based on cooperative network: CI-Net, including a consistency estimation network (C-Net) and inconsistency estimation network (I-Net). C-Net is used to estimate the Main gaze of the true gaze, and an attention mechanism is added to adaptively assign the weight between eyes and face features. The I-Net is used to estimate the Residual residuals based on true gaze. In addition, Cross attention module is designed in this paper, through which I-Net can selectively obtain information from C-Net, to obtain more accurate eyes directions. The experimental results in this paper show that the CI-Net gain lower angle errors than the current mainstream CNN methods under the condition of different appearance of both eyes and facial occlusion.https://ieeexplore.ieee.org/document/9841577/Gaze estimationdeep learningmain gazeresidual residuals |
| spellingShingle | Yuan Luo Jiangtao Chen Jian Chen CI-Net: Appearance-Based Gaze Estimation via Cooperative Network IEEE Access Gaze estimation deep learning main gaze residual residuals |
| title | CI-Net: Appearance-Based Gaze Estimation via Cooperative Network |
| title_full | CI-Net: Appearance-Based Gaze Estimation via Cooperative Network |
| title_fullStr | CI-Net: Appearance-Based Gaze Estimation via Cooperative Network |
| title_full_unstemmed | CI-Net: Appearance-Based Gaze Estimation via Cooperative Network |
| title_short | CI-Net: Appearance-Based Gaze Estimation via Cooperative Network |
| title_sort | ci net appearance based gaze estimation via cooperative network |
| topic | Gaze estimation deep learning main gaze residual residuals |
| url | https://ieeexplore.ieee.org/document/9841577/ |
| work_keys_str_mv | AT yuanluo cinetappearancebasedgazeestimationviacooperativenetwork AT jiangtaochen cinetappearancebasedgazeestimationviacooperativenetwork AT jianchen cinetappearancebasedgazeestimationviacooperativenetwork |