R-CNN-Based Chess Position Recognition

Key points

• 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.1
• For piece recognition, we adapt Faster R-CNN using a class-weighted objective and a deeper classification head.1
• For board detection, we introduce an octagonal arrangement of eight labelled boundary keypoints, predicted using the keypoint head of Mask R-CNN.1
• Using ground-truth piece boxes with the predicted homographies gives correct square assignments for every test position.1

Sources (1)

• [1]R-CNN-Based Chess Position Recognition
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 6, 10:42 PM

  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.

  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.

Extractive summary: sentences quoted from the sources.