BROWSE
Detail
Conference Paper
HAZE-Net: High-Frequency Attentive Super-Resolved Gaze Estimation in Low-Resolution Face Images
Cited 0 time in 
Share 
Share
- Authors
- Jun-Seok Yun, Youngju Na, Hee Hyeon Kim, Hyung-Il Kim, Seok Bong Yoo
- Issue Date
- 2022-12
- Citation
- Asian Conference on Computer Vision (ACCV) 2022, pp.3361-3378
- Language
- English
- Type
- Conference Paper
- Abstract
- Although gaze estimation methods have been developed with deep learning techniques, there has been no such approach as aim to attain accurate performance in low-resolution face images with a pixel width of 50 pixels or less. To solve a limitation under the challenging low-resolution conditions, we propose a high-frequency attentive super-resolved gaze estimation network, i.e., HAZE-Net. Our network improves the resolution of the input image and enhances the eye features and those boundaries via a proposed super-resolution module based on a high-frequency attention block. In addition, our gaze estimation module utilizes high-frequency components of the eye as well as the global appearance map. We also utilize the structural location information of faces to approximate head pose. The experimental results indicate that the proposed method exhibits robust gaze estimation performance even in low-resolution face images with 2828 pixels. The source code of this work is available at https://github.com/dbseorms16/HAZE_Net/.
- KSP Keywords
- Face image, Gaze Estimation, Global appearance, Head pose, High frequency(HF), Location information, Source Code, Super resolution, deep learning(DL), estimation method, estimation performance