Precise event spotting (PES) aims to recognize fine-grained events at exact moments and has become a key component of sports analytics. This task is particularly challenging due to rapid succession, motion blur, and subtle visual differences. Consequently, most existing methods rely on domain-specific, end-to-end training with large labeled datasets and often struggle in few-shot conditions due to their dependence on pixel- or pose-based inputs alone. However, obtaining large labeled datasets is practically hard. We propose a Unified Multi-Entity Graph Network (UMEG-Net) for few-shot PES. UMEG-Net integrates human skeletons and sport-specific object keypoints into a unified graph and features an efficient spatio-temporal extraction module based on advanced GCN and multi-scale temporal shift. To further enhance performance, we employ multimodal distillation to transfer knowledge from keypoint-based graphs to visual representations. Our approach achieves robust performance with limited labeled data and significantly outperforms baseline models in few-shot settings, providing a scalable and effective solution for few-shot PES.

Analyzing Fast, Frequent, and Fine-grained ($F^3$) events presents a significant challenge in video analytics and multi-modal LLMs. Current methods struggle to identify events that satisfy all the $F^3$ criteria with high accuracy due to challenges such as motion blur and subtle visual discrepancies. To advance research in video understanding, we introduce $F^3Set$, a benchmark that consists of video datasets for precise $F^3$ event detection. Datasets in $F^3Set$ are characterized by their extensive scale and comprehensive detail, usually encompassing over 1,000 event types with precise timestamps and supporting multi-level granularity. Currently, $F^3Set$ contains several sports datasets, and this framework may be extended to other applications as well. We evaluated popular temporal action understanding methods on $F^3Set$, revealing substantial challenges for existing techniques. Additionally, we propose a new method, $F^3ED$, for $F^3$ event detections, achieving superior performance. The dataset, model, and benchmark code are available at https://github.com/F3Set/F3Set.

Nowadays there are a wealth of devices and cameras at sports venues and facilities that collect different forms of data. Mining useful insights from such data are crucial for improving the performance of professional athletes. In this paper, we introduce a new interactive tennis analytics framework that can realistically simulate tennis matches using parameters mined from past match data and help reveal in-depth knowledge about tennis strategies. Our approach uses probabilistic model checking to formally evaluate the effectiveness of various strategies and tactics and recommend the best ones for improving players’ chances of winning. Our framework is easily understandable and actionable by players and coaches at any level. We have performed evaluations on tennis matches over the past decade to show the effectiveness of our strategy analytics framework.

Sports analytics encompasses the use of data science, AI, psychology, and IoT devices to improve sports performance, strategy, and decision-making. It involves collecting, processing, and interpreting data from various sources such as video recordings and scouting reports. The data is used to evaluate player and team performance, prevent injuries, and help coaches make informed decisions in game and training. We adopt Probabilistic Model Checking (PMC), a method commonly used in reliability analysis for complex safety systems, and explain how this method can be applied to sports strategy analytics to increase the chance of winning by taking into account the reliability of a player’s specific sub-skill sets. This paper describes how we have integrated PMC, machine learning, and computer vision to develop a new and complex system for sports strategy analytics. Finally, we discuss the vision of a new series of international sports analytics conferences (https://formal-analysis.com/isace/2023/).

Court detection is an important step in sports video analytics. Its goal is to find a projective transformation matrix which projects a standard court to the one visible in the video frame. The inverse of the transformation matrix allows the pixels in the video frame to be projected back to the standard court, hence, provides measurements in the physical world coordinates. Instead of finding the transformation matrix directly, we design a deep fully convolutional neural network to estimate a Perspective Fields (PF) between the court in the source image and the court in a reference image. Our network has two branches, one outputs the PF, the other outputs the court lines which serves as the mask to guide the network to focus on the relevant parts of the input images, thereby finding the transformation pertaining to the court only. Our network is sport-agnostic, which means it is easily generalizable to different types of sports.

Detecting temporally precise and fine-grained events from tennis videos is important in automatic video annotation. This paper addresses the challenges of recognizing a sequence of events from tennis video clips, focusing on accurate timestep identification and distinguishing subtle class differences. We propose a novel but simple end-to-end event detection network to accurately detect and identify the key events, which can be trained on a single GPU. We demonstrate that our model outperforms the existing baselines on our fine-grained tennis event dataset. The research contributes to the development of tennis video analytics and has broader implications in other sports domains.

Each week, there are 2-3 top-level professional tennis tournaments around the world offering attractive prize money and ranking points for the players. Even though match playing is an important part of the professional tennis player’s career development, sports injuries will happen if the player plays too many tournaments. By analyzing historical data and studying cases where players retired during matches, we have been able to identify specific scheduling patterns that correlate with a higher risk of injury. As a result, we propose a dynamic tennis injury prediction model as well as an explainable pattern-mining method to obtain player-specific injury patterns based on historical data. Both approaches take into account features such as the player’s physical condition, match performance, injury history, and recent tournament schedule. With these comprehensive approaches, we aim to balance competitive success and the preservation of players’ well-being.

Decision Making forms the basis of any successful athlete's career. The importance of making the right decision at crucial junctures during a game is eventually the difference between winning or losing a match. Team sports have an important characteristic in that they are usually a set of non-discrete events and a good decision by one athlete, doesn't necessarily lead to a reward. It is a sequence of good decisions that finally lead to the reward, which explains why goals are usually so rare in a game of Soccer. Through this paper, we explore Soccer and look at the number of possibilities that are available at any given time to a player and what is their best possible action. We propose a Model-driven Domain specific Sequential model that aims to predict the best possible action that can be taken by a Soccer Athlete (player), to maximize the probability of scoring a goal.

We are in the era that sport is increasingly defined by data. Rich data is a powerful enabling foundation for novel insights on the game and on player actions, and consequently used for fan engagement and better decision making. Detailed data that has specifications of each shot is currently missing in tennis, as the game dynamics is fast and it is beyond human ability to manually record all specifications of each and every shot, even using scoring software. In this paper, we present an intelligent system to automatically recognize player actions, ball and player movements, and important game events. The system annotates the video with a suitable set of keywords for fast retrieval in broadcast production. Various techniques of computer vision, alignment, filtering, and pre-trained deep learning models are utilized by our system. The evaluation of our results on multiple broadcast videos show great accuracy and timeliness. The implications of the work presented in this paper are profound in the current workflow in broadcast coverage of a tennis match where normally multiple video operators and judges are needed to identify events and retrieve the related clips from multiple cameras.

The advance of analytics technology has attracted more attention and adoption from sports, although modeling and analyzing the dynamic (and uncertain) behaviors of sports are challenging. Formal methods have been strongly recommended to deal with complex systems by their rigorous semantics and powerful reasoning capabilities. In this paper, we present our initiative as the first to apply probabilistic model checking techniques to strategy analytics for tennis based on Markov Decision Processes (MDP). Our approach can derive insights such as prediction of winning chances and identification of best improvement. We evaluate the effectiveness of our approach through real-life case study.