Unsupervised Embedding Learning With Mutual-Information Graph Convolutional Networks

• Published in: IEEE transactions on multimedia Vol. 25; pp. 5916 - 5926
• Main Authors: Zhang, Lin, Zhang, Mingxin, Song, Ran, Zhao, Ziying, Li, Xiaolei
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 2023; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Artificial neural networks; Computer vision; Data mining; Embedding; embedding learning; graph neural network; Image retrieval; Learning; Mutual information; Perturbation; Task analysis; Three-dimensional displays; Training; Unsupervised learning
• ISSN: 1520-9210; 1941-0077
• DOI: 10.1109/TMM.2022.3200852

Recently, methods for unsupervised embedding learning have exhibited promising results for extracting desirable representations from unlabeled samples. In general, most methods learn the feature embeddings by handling each sample individually while the structural and semantic relationships between samples are not fully exploited. As a result, the learned embeddings are not sufficiently discriminative. To make use of such inter-sample information for deep embedding learning, this paper proposes an unsupervised method based on the graph convolutional network (GCN). On one hand, our method encodes structural information between the samples corresponding to the nodes in a local neighbourhood of the GCN graph. On the other hand, it leverages the mutual information between the original samples and the augmented ones to ensure that they are globally consistent with each other. Extensive experiments show that our method is not just robust to augmentation perturbations, but also learns discriminative embeddings. Consequently, it achieves the state-of-the-art performance on several challenging datasets.