TinyFace is a large-scale face recognition benchmark that helps study large-scale native LRFR (Low Resolution Face Recognition) in deep learning.

The dataset has 169,403 native low-resolution face images (average 20×16 pixels) from 5,139 annotated identities for 1:N recognition testing. All low-resolution faces in the dataset are collected from public web data, which are collected under uncontrolled conditions of pose, lighting, occlusion, and background.