TopoFR: A Closer Look at Topology Alignment on Face Recognition

TL;DR

TopoFR introduces a topological structure alignment method using persistent homology and a hard sample mining strategy to enhance face recognition performance and generalization by effectively preserving structure information.

Contribution

The paper proposes TopoFR, a novel face recognition model that leverages persistent homology for topological alignment and a new hard sample mining strategy to improve robustness and accuracy.

Findings

TopoFR outperforms state-of-the-art methods on popular face benchmarks.

Persistent homology effectively preserves topological structure in face recognition.

Hard sample mining improves model robustness against challenging samples.

Abstract

The field of face recognition (FR) has undergone significant advancements with the rise of deep learning. Recently, the success of unsupervised learning and graph neural networks has demonstrated the effectiveness of data structure information. Considering that the FR task can leverage large-scale training data, which intrinsically contains significant structure information, we aim to investigate how to encode such critical structure information into the latent space. As revealed from our observations, directly aligning the structure information between the input and latent spaces inevitably suffers from an overfitting problem, leading to a structure collapse phenomenon in the latent space. To address this problem, we propose TopoFR, a novel FR model that leverages a topological structure alignment strategy called PTSA and a hard sample mining strategy named SDE. Concretely, PTSA uses persistent homology to align the topological structures of the input and latent spaces, effectively preserving the structure information and improving the generalization performance of FR model. To mitigate the impact of hard samples on the latent space structure, SDE accurately identifies hard samples by automatically computing structure damage score (SDS) for each sample, and directs the model to prioritize optimizing these samples. Experimental results on popular face benchmarks demonstrate the superiority of our TopoFR over the state-of-the-art methods. Code and models are available at: https://github.com/modelscope/facechain/tree/main/face_module/TopoFR.