Application of integral invariants to apictorial jigsaw puzzle assembly

TL;DR

This paper introduces a novel method for automatically assembling apictorial jigsaw puzzles using integral area invariants for shape matching, eliminating the need for assumptions about piece shape or arrangement.

Contribution

The paper presents a new approach leveraging integral invariants and optimization for puzzle assembly without prior shape assumptions.

Findings

Successfully assembled various puzzle shapes and sizes

Demonstrated robustness of shape matching method

Achieved automatic puzzle assembly without shape assumptions

Abstract

We present a method for the automatic assembly of apictorial jigsaw puzzles. This method relies on integral area invariants for shape matching and an optimization process to aggregate shape matches into a final puzzle assembly. Assumptions about individual piece shape or arrangement are not necessary. We illustrate our method by solving example puzzles of various shapes and sizes.