Ganzzle: Reframing Jigsaw Puzzle Solving as a Retrieval Task using a Generative Mental Image

Puzzle solving is a combinatorial challenge due to the difficulty of matching adjacent pieces. Instead, we infer a mental image from all pieces, which a given piece can then be matched against avoiding the combinatorial explosion. Exploiting advancements in Generative Adversarial methods, we learn h...

Full description

Saved in:
Bibliographic Details
Main Authors: Talon, Davide, Bue, Alessio Del, James, Stuart
Format: Conference Proceeding
Language:English
Subjects:
Cognition
Deep learning
Explosions
Generative Adversarial Network
Generators
Hungarian Algorithm
Image coding
Image reconstruction
Jigsaw puzzle
Spatial Reasoning
Task analysis
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Puzzle solving is a combinatorial challenge due to the difficulty of matching adjacent pieces. Instead, we infer a mental image from all pieces, which a given piece can then be matched against avoiding the combinatorial explosion. Exploiting advancements in Generative Adversarial methods, we learn how to reconstruct the image given a set of unordered pieces, allowing the model to learn a joint embedding space to match an encoding of each piece to the cropped layer of the generator. Therefore we frame the problem as a R@1 retrieval task, and then solve the linear assignment using differentiable Hungarian attention, making the process end-to-end. In doing so our model is puzzle size agnostic, in contrast to prior deep learning methods which are single size. We evaluate on two new large-scale datasets, where our model is on par with deep learning methods, while generalizing to multiple puzzle sizes.
ISSN:2381-8549
DOI:10.1109/ICIP46576.2022.9897553