4.8 Article

STAR-FC: Structure-Aware Face Clustering on Ultra-Large-Scale Graphs

出版社

IEEE COMPUTER SOC
DOI: 10.1109/TPAMI.2023.3299263

关键词

Training; Faces; Face recognition; Training data; Standards; Clustering algorithms; Task analysis; Face clustering; large-scale graph; graph convolutional network; hierarchical GCN training; node intimacy

向作者/读者索取更多资源

This paper proposes a new face clustering method, called STructure-AwaRe Face Clustering (STAR-FC), which addresses the dilemma of large-scale training and efficient inference by designing a structure-preserving subgraph sampling strategy and a novel hierarchical GCN training paradigm. During inference, the STAR-FC performs efficient full-graph clustering with two steps: graph parsing and graph refinement, and introduces the concept of node intimacy to mine the local structural information. The experimental results demonstrate that this method achieves superior performance and efficiency.
Face clustering is a promising method for annotating unlabeled face images. Recent supervised approaches have boosted the face clustering accuracy greatly, however their performance is still far from satisfactory. These methods can be roughly divided into global-based and local-based ones. Global-based methods suffer from the limitation of training data scale, while local-based ones are inefficient for inference due to the use of numerous overlapped subgraphs. Previous approaches fail to tackle these two challenges simultaneously. To address the dilemma of large-scale training and efficient inference, we propose the STructure-AwaRe Face Clustering (STAR-FC) method. Specifically, we design a structure-preserving subgraph sampling strategy to explore the power of large-scale training data, which can increase the training data scale from 10(5) to 10(7). On this basis, a novel hierarchical GCN training paradigm is further proposed for better capturing the dynamic local structure. During inference, the STAR-FC performs efficient full-graph clustering with two steps: graph parsing and graph refinement. And the concept of node intimacy is introduced in the second step to mine the local structural information, where a calibration module is further proposed for fairer edge scores. The STAR-FC gets 93.21 pairwise F-score on standard partial MS1M within 312 seconds, which far surpasses the state-of-the-arts while maintaining high inference efficiency. Furthermore, we are the first to train on an ultra-large-scale graph with 20 M nodes, and achieve superior inference results on 12 M testing data. Overall, as a simple and effective method, the proposed STAR-FC provides a strong baseline for large-scale face clustering.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.8
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据