ShuttleNet: Position-Aware Fusion of Rally Progress and Player Styles for Stroke Forecasting in Badminton

Abstract

The increasing demand for analyzing the insights in sports has stimulated a line of productive studies from a variety of perspectives, e.g., health state monitoring, outcome prediction. In this paper, we focus on objectively judging what and where to return strokes, which is still unexplored in turn-based sports. By formulating stroke forecasting as a sequence prediction task, existing works can tackle the problem but fail to model information based on the characteristics of badminton. To address these limitations, we propose a novel Position-aware Fusion of Rally Progress and Player Styles framework (ShuttleNet) that incorporates rally progress and information of the players by two modified encoder-decoder extractors. Moreover, we design a fusion network to integrate rally contexts and contexts of the players by conditioning on information dependency and different positions. Extensive experiments on the badminton dataset demonstrate that ShuttleNet significantly outperforms the state-of-the-art methods and also empirically validates the feasibility of each component in ShuttleNet. On top of that, we provide an analysis scenario for the stroke forecasting problem.

ShuttleSet: A Human-Annotated Stroke-Level Singles Dataset for Badminton Tactical Analysis

“…In this section, we present 3 application scenarios related to the performance on ShuttleSet, including measuring the win probability of each stroke in a rally (shot influence [41,42], Section 5.1), forecasting the future strokes (stroke forecasting [43], Section 5.2) and the movements of both players (movement forecasting [3], Section 5.3) given the previous strokes. Following their evaluation settings for data separation and evaluation metrics, the shot influence task formulates the task as predicting the final outcome of a rally and adopts the latest 10 matches as the test set and the remaining for the train set to evaluate model effectiveness for inferring the win probability of each stroke in the unseen matches.…”

“…In addition to the classification task, we also implement forecasting tasks using ShuttleSet, which is generally more challenging than the classification task. Wang et al [43] proposes the stroke forecasting task, which is beneficial for not only simulating players' tactics but also assessing the returning probability of future strokes for storytelling. This task is defined as predicting future strokes including shot types and the corresponding destination locations based on the previous strokes in a rally.…”

“…This task is defined as predicting future strokes including shot types and the corresponding destination locations based on the previous strokes in a rally. We reproduced the state-of-the-art model, ShuttleNet [43], and the baselines Seq2Seq [37], CF-LSTM [46], TF [13], dNRI [14], and DMA-Nets [20] for illustrating the performance of shot types and landing locations. Seq2Seq uses two long short-term memory (LSTM) [33] as an encoder and a decoder.…”

“…For example, a video assistant referee system has been deployed in football over 100 competitions including the 2022 FIFA World Cup to support the referee team to reduce misjudgments 1 . On the other hand, player evaluation and tactic investigation have improved the effectiveness in multiple sports, such as football [1,7,23,36], baseball [16,24], basketball [4,47], and badminton [3,43]. To train models for these applications, the datasets are mainly collected from either videos [9,50] or sensors [15,18,29] to extract the enormous variable metadata in the matches.…”

A Survey on Video Action Recognition in Sports: Datasets, Methods and Applications

“…Stroke Forecasting [210] is a most recent dataset, consisting of 43,191 trimmed video clips and each video clip has a stroke belongs to one of 10 categories -smash, push, clear, defensive shot, net shot, drive, drop, lob, long service and short service. In addition to badminton action recognition, the dataset can also be used for stroke forecasting, i.e., given previous stokes in a rally, the model should predict what the next stroke is.…”