R-CNN-Based Chess Position Recognition
Quick Answer
This paper shows that An R-CNN-based framework enhances chess position recognition from images by integrating piece recognition and board geometry estimation.
Quick Take
It achieves a mean average precision of 90.14% for piece detection and accurately recovers 76.61% of test positions, with 96.49% having at most one incorrect square.
Key Points
- Utilizes Faster R-CNN with a class-weighted objective for piece recognition.
- Introduces octagonal keypoints for robust board detection and orientation estimation.
- Achieves 90.14% mean average precision in piece detection on synthetic datasets.
- 97.11% of predicted board keypoints are within 1% of their labeled targets.
- 76.61% of test positions are recovered exactly with high accuracy.
Paper Resources
📖 Reader Mode
~2 min readAbstract:Performing chess game position recognition solely from a single image of a three-dimensional board requires predicting the position and orientation of the board relative to the camera, the occupancy of squares and the piece type, which includes its colour. We propose an R-CNN-based framework with independent components for piece recognition and board geometry estimation, whose predictions are combined to reconstruct the position. For piece recognition, we adapt Faster R-CNN using a class-weighted objective and a deeper classification head. The detector operates directly on the input image, retaining alternative piece hypotheses that are subsequently refined using constraints on piece counts and square occupancy. For board detection, we introduce an octagonal arrangement of eight labelled boundary keypoints, predicted using the keypoint head of Mask R-CNN. These provide redundant correspondences for homography estimation and encode board orientation. The estimated homography maps representative points from the piece boxes to an 8x8 grid. On a synthetic dataset, the modifications to piece detection increase mean average precision from 61.59% to 90.14%. Of the predicted board keypoints, 97.11% are within 1% of the image diagonal of their labelled targets. Using ground-truth piece boxes with the predicted homographies gives correct square assignments for every test position. The complete framework recovers 76.61% of test positions exactly and 96.49% with at most one incorrect square.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2610.09191 [cs.CV] |
| (or arXiv:2610.09191v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.09191 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Ognjen Arandjelović PhD [view email]
[v1]
Tue, 6 Oct 2026 22:42:22 UTC (15,935 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →ProMoE-FL: Prototype-conditioned Mixture of Experts for Multimodal Federated Learning with Missing Modalities
ProMoE-FL introduces a Prototype-conditioned Mixture-of-Experts framework for multimodal federated learning, effectively addressing missing modalities. It outperforms existing methods on four chest X-ray datasets, demonstrating superior feature synthesis capabilities in both homogeneous and heterogeneous settings.