Agglomerative Hierarchical Clustering based Method for Image Clustering using Restricted Boltzmann Machine

In various verification systems, Restricted Boltzmann Machines (RBMs) have demonstrated their efficacy in both the front-end and back-end processes. In this study, we introduce the application of RBMs to the front-end for image clustering tasks. RBMs are trained to convert images into vector-based representations. In this work, we employ the conventional bottom-up Agglomerative Hierarchical Clustering (AHC) technique. To address the challenge of limited test face image data, we propose RBM adaptation to a global model. Initially, a universal RBM model is trained using all available training data. Subsequently, we train an adapted RBM model using the data from each test image. The visible-to-hidden weight matrices of these adapted models, along with the bias vectors, are concatenated and subjected to whitening to generate vector representations of face images, known as RBM vectors. These vectors effectively preserve class-specific information and are utilized in image clustering tasks. Our experimental results, conducted on two benchmark image datasets (MS-Celeb-1M and DeepFashion), demonstrate that our proposed approach surpasses well-known clustering algorithms such as k-means, spectral clustering, and approximate Rank-order.