Unsupervised Image Sorting

Introduction

Figure 1: Overview of our approach. An unsorted database of images (up) and an output of the proposed method (down)

The goal of this work is to approximately solve the problem of unsupervised image sorting that is considered as a kind of content-based image clustering. The content-based image sorting is the creation of a route that passes through all the images once, in such an order that the next one from the previous image has similar content.

In the end, an image ordering (e.g. slideshow) is automatically produced, so that the images with similar content should be close to each other. This problem resembles the problem known in the literature as ‘travelling salesman problem’ (TSP).

Two classes of methods (the nearest-neighbour and genetic methods) has been proposed that have also been applied on the TSP problem.

We have modified three TSP-based methods: the NN, the NN1 and the genetic methods to solve the image sorting problem using CLD for image representation [1].

Their benefits on computational efficiency and accuracy are discussed over six datasets that have been created from the GHIM-10K dataset. The experimental results demonstrate that the proposed methods efficiently solve the image sorting problem, producing image sequences that almost agree with human intuition.

Experiments - Downloads

    • You can download the matlab code (.rar) of the methods proposed in [1]
    • You can download the datasets of [1] datasets (.rar)
    • See the corresponding readme.txt files for more details.

Related Publications

[1] S. Markaki, C. Panagiotakis, D. Lasthiotaki, Image Sorting via a Reduction to Traveling Salesman Problem, IET Image Processing, 2019.