Learning to Cluster Faces via Confidence and Connectivity Estimation

TL;DR

This paper introduces a fully learnable face clustering framework using graph convolutional networks to estimate vertex confidence and edge connectivity, significantly improving accuracy and efficiency over prior supervised methods.

Contribution

It presents a novel clustering approach that replaces heuristic steps with learned confidence and connectivity estimation via GCNs, eliminating the need for many overlapped subgraphs.

Findings

Significantly improves clustering accuracy on large-scale benchmarks.

Enhances face recognition model performance trained on clustered data.

Achieves an order of magnitude better efficiency than existing supervised methods.

Abstract

Face clustering is an essential tool for exploiting the unlabeled face data, and has a wide range of applications including face annotation and retrieval. Recent works show that supervised clustering can result in noticeable performance gain. However, they usually involve heuristic steps and require numerous overlapped subgraphs, severely restricting their accuracy and efficiency. In this paper, we propose a fully learnable clustering framework without requiring a large number of overlapped subgraphs. Instead, we transform the clustering problem into two sub-problems. Specifically, two graph convolutional networks, named GCN-V and GCN-E, are designed to estimate the confidence of vertices and the connectivity of edges, respectively. With the vertex confidence and edge connectivity, we can naturally organize more relevant vertices on the affinity graph and group them into clusters. Experiments on two large-scale benchmarks show that our method significantly improves clustering accuracy and thus performance of the recognition models trained on top, yet it is an order of magnitude more efficient than existing supervised methods.