Image Reassembly Combining Deep Learning and Shortest Path Problem

TL;DR

This paper introduces a novel approach to image reassembly from fragments by combining deep learning for predicting fragment positions with shortest path algorithms, supported by a new MET dataset and multiple algorithms.

Contribution

It proposes new neural architectures for fragment position prediction, formulates reassembly as a shortest path problem, and introduces a dedicated MET dataset with baseline results.

Findings

Deep neural architectures outperform previous methods.

Graph-based shortest path algorithms effectively reassemble images.

New MET dataset enables standardized evaluation.

Abstract

This paper addresses the problem of reassembling images from disjointed fragments. More specifically, given an unordered set of fragments, we aim at reassembling one or several possibly incomplete images. The main contributions of this work are: 1) several deep neural architectures to predict the relative position of image fragments that outperform the previous state of the art; 2) casting the reassembly problem into the shortest path in a graph problem for which we provide several construction algorithms depending on available information; 3) a new dataset of images taken from the Metropolitan Museum of Art (MET) dedicated to image reassembly for which we provide a clear setup and a strong baseline.