DeepDarts: Modeling Keypoints as Objects for Automatic Scorekeeping in Darts using a Single Camera
Darts are usually played as a traditional pub game. The player here must keep his own score, which slows down the game and makes it less enjoyable. Some automated systems have been created for this end, but they are expensive and sophisticated.
A recent paper proposes a deep learning-based method to predict dart scores from a single image taken from any front-view camera angle.
The researchers suggest a new deep learning-based approach to keypoint detection in which keypoints are modeled as objects. A deep convolutional neural network is used to detect dartboard calibration points in addition to the dart landing positions.
The calibration points are used to map the predicted dart locations and calibrate the scoring area. The scores are then classified based on their position. The experiments confirm that the suggested method can predict dart scores precisely and generalizes to various camera angles.
Existing multi-camera solutions for automatic scorekeeping in steel-tip darts are very expensive and thus inaccessible to most players. Motivated to develop a more accessible low-cost solution, we present a new approach to keypoint detection and apply it to predict dart scores from a single image taken from any camera angle. This problem involves detecting multiple keypoints that may be of the same class and positioned in close proximity to one another. The widely adopted framework for regressing keypoints using heatmaps is not well-suited for this task. To address this issue, we instead propose to model keypoints as objects. We develop a deep convolutional neural network around this idea and use it to predict dart locations and dartboard calibration points within an overall pipeline for automatic dart scoring, which we call DeepDarts. Additionally, we propose several task-specific data augmentation strategies to improve the generalization of our method. As a proof of concept, two datasets comprising 16k images originating from two different dartboard setups were manually collected and annotated to evaluate the system. In the primary dataset containing 15k images captured from a face-on view of the dartboard using a smartphone, DeepDarts predicted the total score correctly in 94.7% of the test images. In a second more challenging dataset containing limited training data (830 images) and various camera angles, we utilize transfer learning and extensive data augmentation to achieve a test accuracy of 84.0%. Because DeepDarts relies only on single images, it has the potential to be deployed on edge devices, giving anyone with a smartphone access to an automatic dart scoring system for steel-tip darts. The code and datasets are available.
Research paper: McNally, W., Walters, P., Vats, K., Wong, A., and McPhee, J., “DeepDarts: Modeling Keypoints as Objects for Automatic Scorekeeping in Darts using a Single Camera”, 2021. Link: https://arxiv.org/abs/2105.09880