Genetic Algorithm-Based Solver for Very Large Multiple Jigsaw Puzzles of Unknown Dimensions and Piece Orientation

TL;DR

This paper introduces a genetic algorithm-based method for solving very large, complex jigsaw puzzles with unknown dimensions and piece orientations, achieving state-of-the-art accuracy and size handling.

Contribution

It presents the first GA-based solver capable of handling unknown puzzle sizes and orientations, with a novel crossover technique and simultaneous multi-puzzle assembly.

Findings

Sets new state-of-the-art in puzzle size and accuracy

Successfully assembles multiple puzzles simultaneously

Outperforms previous methods even with unknown dimensions

Abstract

In this paper we propose the first genetic algorithm (GA)-based solver for jigsaw puzzles of unknown puzzle dimensions and unknown piece location and orientation. Our solver uses a novel crossover technique, and sets a new state-of-the-art in terms of the puzzle sizes solved and the accuracy obtained. The results are significantly improved, even when compared to previous solvers assuming known puzzle dimensions. Moreover, the solver successfully contends with a mixed bag of multiple puzzle pieces, assembling simultaneously all puzzles.