Search

Tennis is a popular sport, and integrating modern technological advancements can greatly enhance player training. Human pose estimation has seen substantial developments recently, driven by progress in deep learning. The dataset described in this paper was compiled from videos of researchers’ friend playing tennis. These videos were retrieved frame by frame to categorize various tennis movements, and human skeleton joints were annotated using COCO-Annotator to generate labelled JSON files. By combining these JSON files with the classified image set, we constructed the dataset for this paper. This dataset enables the training and validation of four tennis postures, forehand shot, backhand shot, ready position, and serves, using deep learning models (such as OpenPose). The researchers believe that this dataset will be a valuable asset to the tennis community and human pose estimation field, fostering innovation and excellence in the sport.

Tennis is a popular sport, and the introduction of technology has allowed players to diversify their training. Tennis ball tracking is currently a focal point, serving not only to assist referees but also to enhance sports analysis. We introduce the Tennis Shot Side-View and Top-View Dataset, which serves as an invaluable resource for analyzing tennis movements and verifying landing positions after flight. This dataset combines side-view and top-view video clips, capturing various shot types and player movements from both outdoor and indoor fields. The dataset includes the actual ball positions of each clip for verification purposes. The Tennis Shot Side-View and Top-View Dataset represents a significant advancement in tennis research. Its multidimensional nature opens doors for in-depth player analysis, performance enhancement, and strategy development. We believe that this dataset will be a valuable asset to the tennis community, fostering innovation and excellence in the sport.