# Improving Deep Image Clustering With Spatial Transformer Layers

## TL;DR

This paper enhances deep image clustering by integrating Spatial Transformer Networks with Deep Adaptive Clustering, effectively addressing spatial transformations like scale and rotation, and demonstrating improved performance on MNIST datasets.

## Contribution

It introduces a novel combination of visual attention techniques with deep clustering models to better handle spatial variations in images.

## Key findings

- Outperforms baseline models on MNIST and FashionMNIST datasets.
- Effectively reduces issues caused by spatial transformations.
- Demonstrates the benefit of combining STN with deep clustering.

## Abstract

Image clustering is an important but challenging task in machine learning. As in most image processing areas, the latest improvements came from models based on the deep learning approach. However, classical deep learning methods have problems to deal with spatial image transformations like scale and rotation. In this paper, we propose the use of visual attention techniques to reduce this problem in image clustering methods. We evaluate the combination of a deep image clustering model called Deep Adaptive Clustering (DAC) with the Spatial Transformer Networks (STN). The proposed model is evaluated in the datasets MNIST and FashionMNIST and outperformed the baseline model.