QMUL TinyFace


We create a large scale face recognition benchmark, named TinyFace, to facilitate the investigation of natively LRFR at large scales (large gallery population sizes) in deep learning. The TinyFace dataset consists of 5,139 labelled facial identities given by 169,403 native LR face images (average 20×16 pixels) designed for 1:N recognition test. All the LR faces in TinyFace are collected from public web data across a large variety of imaging scenarios, captured under uncontrolled viewing conditions in pose, illumination, occlusion and background. Beyond artificially down-sampling HR face images for LRFR performance test as in previous works, to our best knowledge, this is the first systematic study focusing specially on face recognition of native LR images.


TinyFace Dataset and Evaluation Codes: (148MB): [Google Drive] [Baidu Cloud]


Please feel free to send any questions and/or comments to Zhiyi Cheng at z.cheng@qmul.ac.uk