MixFaceNets: Extremely Efficient Face Recognition Networks

TL;DR

MixFaceNets are highly efficient face recognition models inspired by mixed depthwise convolutions, achieving state-of-the-art accuracy with extremely low computational cost across multiple face verification benchmarks.

Contribution

Introduction of MixFaceNets, a new set of face recognition models that outperform existing lightweight models at similar computational complexity.

Findings

Outperform MobileFaceNets at < 500M FLOPs across datasets.

Achieve 99.60% accuracy on LFW.

Comparable results to top models at 500M-1G FLOPs.

Abstract

In this paper, we present a set of extremely efficient and high throughput models for accurate face verification, MixFaceNets which are inspired by Mixed Depthwise Convolutional Kernels. Extensive experiment evaluations on Label Face in the Wild (LFW), Age-DB, MegaFace, and IARPA Janus Benchmarks IJB-B and IJB-C datasets have shown the effectiveness of our MixFaceNets for applications requiring extremely low computational complexity. Under the same level of computation complexity (< 500M FLOPs), our MixFaceNets outperform MobileFaceNets on all the evaluated datasets, achieving 99.60% accuracy on LFW, 97.05% accuracy on AgeDB-30, 93.60 TAR (at FAR1e-6) on MegaFace, 90.94 TAR (at FAR1e-4) on IJB-B and 93.08 TAR (at FAR1e-4) on IJB-C. With computational complexity between 500M and 1G FLOPs, our MixFaceNets achieved results comparable to the top-ranked models, while using significantly fewer FLOPs and less computation overhead, which proves the practical value of our proposed MixFaceNets. All training codes, pre-trained models, and training logs have been made available https://github.com/fdbtrs/mixfacenets.