Large-scale Face Clustering Method Research Based on Deep Learning
Zixin Wen · 2021 3rd International Conference on Machine Learning, Big Data and Business Intelligence (MLBDBI) · 2021
Face clustering is a pre-processing stage for face detection and face recognition tasks, alleviating the heavy manual labeling work. The goodness of clustering depends on the data features and clustering algorithm. The current data is so complex and diverse that traditional machine learning clustering algorithms can no longer handle such large-scale and high latitude data. Therefore, this paper introduces two types of face clustering algorithms based on deep learning, face clustering algorithms based on deep subspace and face clustering algorithms based on graph convolutional networks, and introduces each algorithm and compares its advantages and shortcomings.