Iris-LAHNet: a lightweight attention-guided high-resolution network for iris segmentation and localization

Iris recognition models that can be deployed on mobile devices have further requirements for both model scale and accuracy. We note that iris segmentation and localization tasks are the basis of iris recognition. Therefore, to better meet the needs of the community, we propose a lightweight attention-guided high-resolution network (Iris-LAHNet). Iris-LAHNet is composed of a stem, a basic backbone, pyramid dilated convolution (PDC) blocks, cascaded attention-guided feature fusion module (C-AGFM), and auxiliary heads. The basic backbone is a tiny high-resolution network. The introduction of PDC blocks and C-AGFM helps to extract multi-scale features from multi-resolution images and reduce noise. In addition, we introduce three auxiliary heads with edge heatmaps, which output auxiliary loss to help model training and enhance attention to single pixels of the edge. It helps to compensate for the neglect of localization tasks during multi-task training. Experiments on four datasets show that our model achieves the lightest while ensuring segmentation and localization results.