Show simple item record

Files in this item

Thumbnail

Item metadata

dc.contributor.authorWölflein, Georg
dc.contributor.authorArandelovic, Oggie
dc.date.accessioned2021-06-02T14:30:10Z
dc.date.available2021-06-02T14:30:10Z
dc.date.issued2021-06-02
dc.identifier.citationWölflein , G & Arandelovic , O 2021 , ' Determining chess game state from an image ' , Journal of Imaging , vol. 7 , no. 6 , 94 . https://doi.org/10.3390/jimaging7060094en
dc.identifier.issn2313-433X
dc.identifier.otherPURE: 274375214
dc.identifier.otherPURE UUID: 2628f895-a0b0-4910-8cc3-7af45f23a8f3
dc.identifier.urihttp://hdl.handle.net/10023/23299
dc.description.abstractIdentifying the configuration of chess pieces from an image of a chessboard is a problem in computer vision that has not yet been solved accurately. However, it is important for helping amateur chess players improve their games by facilitating automatic computer analysis without the overhead of manually entering the pieces. Current approaches are limited by the lack of large datasets and are not designed to adapt to unseen chess sets. This paper puts forth a new dataset synthesised from a 3D model that is an order of magnitude larger than existing ones. Trained on this dataset, a novel end-to-end chess recognition system is presented that combines traditional computer vision techniques with deep learning. It localises the chessboard using a RANSAC-based algorithm that computes a projective transformation of the board onto a regular grid. Using two convolutional neural networks, it then predicts an occupancy mask for the squares in the warped image and finally classifies the pieces. The described system achieves an error rate of 0.23% per square on the test set, 28 times better than the current state of the art. Further, a few-shot transfer learning approach is developed that is able to adapt the inference system to a previously unseen chess set using just two photos of the starting position, obtaining a per-square accuracy of 99.83% on images of that new chess set. The code, dataset, and trained models are made available online.
dc.format.extent16
dc.language.isoeng
dc.relation.ispartofJournal of Imagingen
dc.rightsCopyright © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/.en
dc.subjectComputer visionen
dc.subjectChessen
dc.subjectConvolutional neural networksen
dc.subjectQA75 Electronic computers. Computer scienceen
dc.subjectDASen
dc.subject.lccQA75en
dc.titleDetermining chess game state from an imageen
dc.typeJournal articleen
dc.description.versionPublisher PDFen
dc.contributor.institutionUniversity of St Andrews.School of Computer Scienceen
dc.identifier.doihttps://doi.org/10.3390/jimaging7060094
dc.description.statusPeer revieweden
dc.identifier.urlhttps://www.mdpi.com/2313-433X/7/6/94en


This item appears in the following Collection(s)

Show simple item record