TransFace: Parametric Attention based Transformer for Face Recognition

Zhaochuan Luo, Zhanjia Yang, Gui Wang, Chenqi Yin, Xianxu Hou, Linlin Shen

Research output: Chapter in Book/Conference proceedingConference contributionpeer-review


Face recognition has made remarkable success thanks to the excellent representational learning of deep Convolutional Neural Networks (CNNs). Recently, Transformer-based methods have achieved extraordinary results in various computer vision tasks, which motivates us to explore efficient Transformer for face recognition. Based on the observation that face patches at the same positions from different identities commonly represent similar semantics, we propose TransFace that improves the self-attention mechanism into parametirc-attention for face recognition. Our model is trained on MSCeleb-1M datasets, and evaluated on several benchmarks, i.e., LFW, CALFW, CPLFW, IJB-B and IJB-C datasets. Extensive experiments demonstrate that the cancellation of Q and K does not cause degradation of performance, and the modified parametric-attention has good portability. It is demonstrated that TransFace can achieve better results than CNNs with the same computational complexity.

Original languageEnglish
Title of host publication7th International Conference on Vision, Image and Signal Processing, ICVISP 2023
PublisherInstitution of Engineering and Technology
Number of pages7
ISBN (Electronic)9781837240135, 9781837240142, 9781837240210, 9781839538551, 9781839538650, 9781839539022, 9781839539091, 9781839539107, 9781839539176, 9781839539220, 9781839539237, 9781839539275, 9781839539305, 9781839539312, 9781839539329, 9781839539350, 9781839539367, 9781839539404, 9781839539497, 9781839539503, 9781839539572, 9781839539596, 9781839539664, 9781839539671, 9781839539824, 9781839539831, 9781839539848, 9781839539916, 9781839539961, 9781839539978, 9781839539985, 9781839539992
Publication statusPublished - 2023
Externally publishedYes
Event7th International Conference on Vision, Image and Signal Processing, ICVISP 2023 - Dali, China
Duration: 24 Nov 202326 Nov 2023


Conference7th International Conference on Vision, Image and Signal Processing, ICVISP 2023


  • Face recognition
  • Parametric-Attention
  • Transformer

ASJC Scopus subject areas

  • General Engineering


Dive into the research topics of 'TransFace: Parametric Attention based Transformer for Face Recognition'. Together they form a unique fingerprint.

Cite this