FaceFusion: Exploiting Full Spectrum of Multiple Datasets

TL;DR

FaceFusion is a novel training approach that combines multiple face datasets into a conflict-free unified view, enabling end-to-end training of face recognition models with improved performance across various benchmarks.

Contribution

The paper introduces FaceFusion, a new method for effectively combining multiple datasets without label conflicts for enhanced face recognition training.

Findings

Outperforms single dataset training in accuracy.

Surpasses previous methods in public benchmarks.

Effective in various training scenarios.

Abstract

The size of training dataset is known to be among the most dominating aspects of training high-performance face recognition embedding model. Building a large dataset from scratch could be cumbersome and time-intensive, while combining multiple already-built datasets poses the risk of introducing large amount of label noise. We present a novel training method, named FaceFusion. It creates a fused view of different datasets that is untainted by identity conflicts, while concurrently training an embedding network using the view in an end-to-end fashion. Using the unified view of combined datasets enables the embedding network to be trained against the entire spectrum of the datasets, leading to a noticeable performance boost. Extensive experiments confirm superiority of our method, whose performance in public evaluation datasets surpasses not only that of using a single training dataset, but also that of previously known methods under various training circumstances.