A Transfer Learning based Feature-Weak-Relevant Method for Image Clustering

TL;DR

This paper introduces a novel image clustering method that combines unsupervised learning with transfer learning in an iterative process, reducing reliance on specific features and improving clustering performance.

Contribution

It proposes a feature-weak-relevant approach that iteratively refines image clusters by integrating handcrafted features, sampling strategies, and transfer learning, outperforming existing methods.

Findings

Outperforms state-of-the-art clustering methods on six datasets.

Reduces dependence on specific features for clustering.

Demonstrates effective iterative optimization of image clusters.

Abstract

Image clustering is to group a set of images into disjoint clusters in a way that images in the same cluster are more similar to each other than to those in other clusters, which is an unsupervised or semi-supervised learning process. It is a crucial and challenging task in machine learning and computer vision. The performances of existing image clustering methods have close relations with features used for clustering, even if unsupervised coding based methods have improved the performances a lot. To reduce the effect of clustering features, we propose a feature-weak-relevant method for image clustering. The proposed method converts an unsupervised clustering process into an alternative iterative process of unsupervised learning and transfer learning. The clustering process firstly starts up from handcrafted features based image clustering to estimate an initial label for every image, and secondly use a proposed sampling strategy to choose images with reliable labels to feed a transfer-learning model to learn representative features that can be used for next round of unsupervised learning. In this manner, image clustering is iteratively optimized. What's more, the handcrafted features are used to boot up the clustering process, and just have a little effect on the final performance; therefore, the proposed method is feature-weak-relevant. Experimental results on six kinds of public available datasets show that the proposed method outperforms state of the art methods and depends less on the employed features at the same time.