ArbGaze: Gaze Estimation from Arbitrary-Sized Low-Resolution Images
The goal of gaze estimation is to estimate a gaze vector from an image containing a face or eye(s). Most existing studies use pre-defined fixed-resolution images to estimate the gaze vector. However, images captured from in-the-wild environments may have various resolutions, and variation in resolut...
| Asıl Yazarlar: | , |
|---|---|
| Materyal Türü: | Makale |
| Dil: | English |
| Baskı/Yayın Bilgisi: |
MDPI AG
2022-09-01
|
| Seri Bilgileri: | Sensors |
| Konular: | |
| Online Erişim: | https://www.mdpi.com/1424-8220/22/19/7427 |
| _version_ | 1827652988800335872 |
|---|---|
| author | Hee Gyoon Kim Ju Yong Chang |
| author_facet | Hee Gyoon Kim Ju Yong Chang |
| author_sort | Hee Gyoon Kim |
| collection | DOAJ |
| description | The goal of gaze estimation is to estimate a gaze vector from an image containing a face or eye(s). Most existing studies use pre-defined fixed-resolution images to estimate the gaze vector. However, images captured from in-the-wild environments may have various resolutions, and variation in resolution can degrade gaze estimation performance. To address this problem, a gaze estimation method from arbitrary-sized low-resolution images is proposed. The basic idea of the proposed method is to combine knowledge distillation and feature adaptation. Knowledge distillation helps the gaze estimator for arbitrary-sized images generate a feature map similar to that from a high-resolution image. Feature adaptation makes creating a feature map adaptive to various resolutions of an input image possible by using a low-resolution image and its scale information together. It is shown that combining these two ideas improves gaze estimation performance substantially in the ablation study. It is also demonstrated that the proposed method can be generalized to other popularly used gaze estimation models through experiments using various backbones. |
| first_indexed | 2024-03-09T21:09:49Z |
| format | Article |
| id | doaj.art-17f6e4a9ac334310b924ce25762b8888 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T21:09:49Z |
| publishDate | 2022-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-17f6e4a9ac334310b924ce25762b88882023-11-23T21:49:01ZengMDPI AGSensors1424-82202022-09-012219742710.3390/s22197427ArbGaze: Gaze Estimation from Arbitrary-Sized Low-Resolution ImagesHee Gyoon Kim0Ju Yong Chang1Department of Electronics and Communications Engineering, Kwangwoon University, Seoul 01897, KoreaDepartment of Electronics and Communications Engineering, Kwangwoon University, Seoul 01897, KoreaThe goal of gaze estimation is to estimate a gaze vector from an image containing a face or eye(s). Most existing studies use pre-defined fixed-resolution images to estimate the gaze vector. However, images captured from in-the-wild environments may have various resolutions, and variation in resolution can degrade gaze estimation performance. To address this problem, a gaze estimation method from arbitrary-sized low-resolution images is proposed. The basic idea of the proposed method is to combine knowledge distillation and feature adaptation. Knowledge distillation helps the gaze estimator for arbitrary-sized images generate a feature map similar to that from a high-resolution image. Feature adaptation makes creating a feature map adaptive to various resolutions of an input image possible by using a low-resolution image and its scale information together. It is shown that combining these two ideas improves gaze estimation performance substantially in the ablation study. It is also demonstrated that the proposed method can be generalized to other popularly used gaze estimation models through experiments using various backbones.https://www.mdpi.com/1424-8220/22/19/7427gaze estimationknowledge distillationfeature adaptationdeep neural network |
| spellingShingle | Hee Gyoon Kim Ju Yong Chang ArbGaze: Gaze Estimation from Arbitrary-Sized Low-Resolution Images Sensors gaze estimation knowledge distillation feature adaptation deep neural network |
| title | ArbGaze: Gaze Estimation from Arbitrary-Sized Low-Resolution Images |
| title_full | ArbGaze: Gaze Estimation from Arbitrary-Sized Low-Resolution Images |
| title_fullStr | ArbGaze: Gaze Estimation from Arbitrary-Sized Low-Resolution Images |
| title_full_unstemmed | ArbGaze: Gaze Estimation from Arbitrary-Sized Low-Resolution Images |
| title_short | ArbGaze: Gaze Estimation from Arbitrary-Sized Low-Resolution Images |
| title_sort | arbgaze gaze estimation from arbitrary sized low resolution images |
| topic | gaze estimation knowledge distillation feature adaptation deep neural network |
| url | https://www.mdpi.com/1424-8220/22/19/7427 |
| work_keys_str_mv | AT heegyoonkim arbgazegazeestimationfromarbitrarysizedlowresolutionimages AT juyongchang arbgazegazeestimationfromarbitrarysizedlowresolutionimages |