Deep Convolutional K-Means Clustering

• Published in: Proceedings - International Conference on Image Processing pp. 211 - 215
• Main Authors: Goel, Anurag, Majumdar, Angshul, Chouzenoux, Emilie, Chierchia, Giovanni
• Format: Conference Proceeding
• Language: English
• Published: IEEE 16.10.2022
• Subjects: Benchmark testing; Convolutional Neural Network; Convolutional Transform Learning; Decoding; Deep learning; Image processing; K-means Clustering; Representation learning; Training; Transforms
• ISSN: 2381-8549
• DOI: 10.1109/ICIP46576.2022.9897742

Conventional Convolutional Neural Network (CNN) based clustering formulations are based on the encoder-decoder based framework, where the clustering loss is incorporated after the encoder network. The problem with this approach is that it requires training an additional decoder network; this, in turn, means learning additional weights which can lead to over-fitting in data constrained scenarios. This work introduces a Deep Convolutional Transform Learning (DCTL) based clustering framework. The advantage of our proposed formulation is that we do not require learning the additional decoder network. Therefore our formulation is less prone to over-fitting. Comparison with state-of-the-art deep learning based clustering solutions on benchmark image datasets shows that our proposed method improves over the rest in challenging scenarios where there are many clusters with limited samples.