About this document
ELANet: Efficient Face Recognition Network by divevivi334.kor is a document available to read on EtoBox.
This document describes an efficient lightweight attention network called ELANet for face recognition. ELANet uses spatial attention to capture important local patches and channel attention to focus on important features. It also uses an Efficient Fusion Attention module to fuse spatial and channel attention with less computation. Additionally, ELANet employs a pyramid multi-scale module to learn multi-scale features and an Adaptively Spatial Feature Fusion module to fuse features from different layers. Eva
- Author
- divevivi334.kor
- Language
- EN