Automated recognition of the cricket batting backlift technique in video footage using deep learning architectures

Abstract

There have been limited studies demonstrating the validation of batting techniques in cricket using machine learning. This study demonstrates how the batting backlift technique in cricket can be automatically recognised in video footage and compares the performance of popular deep learning architectures, namely, AlexNet, Inception V3, Inception Resnet V2, and Xception. A dataset is created containing the lateral and straight backlift classes and assessed according to standard machine learning metrics. The architectures had similar performance with one false positive in the lateral class and a precision score of 100%, along with a recall score of 95%, and an f1-score of 98% for each architecture, respectively. The AlexNet architecture performed the worst out of the four architectures as it incorrectly classified four images that were supposed to be in the straight class. The architecture that is best suited for the problem domain is the Xception architecture with a loss of 0.03 and 98.2.5% accuracy, thus demonstrating its capability in differentiating between lateral and straight backlifts. This study provides a way forward in the automatic recognition of player patterns and motion capture, making it less challenging for sports scientists, biomechanists and video analysts working in the field.

Analysis of Upper Body Three-D Motion of Batters for Execution of Pull Shot in Cricket

“…One of the camera was positioned 13 meters in front of batter, and the other was installed to side nearer to bowling crease. Researched focused a 24-point aluminum calibration frame for capturing video volume (Moodley et al, 2022). Calibration volume was set at 3-meters on the X-axis for sagittal plane motions, 1.5-meters on Y-axis for frontal plane, and 2-meters on Z-axis in vertical position for sloping plane motions (Bagchi, 2014).…”

“…Analysis of variance with repeated measures to examine the three levels (stance, back lift, & impact) with two situations (successful vs. unsuccessful pull-shot) were used. Nature of data such as normality, the homogeneity of the variance, and multi-collinear assumptions were properly verified (Moodley et al, 2022). Level of significance was kept as P < .05 and the significance of the difference between the successful and unsuccessful pull shots was calculated.…”

“…The pull shot in cricket is played with swing of bat in horizontal position to take maximum benefit of short-pitched delivery and smash it in most suitable direction (Semwal, Mishra, Raj, Sharma & Mittal, 2018). In cricket pull shot the upper arm elbow position is important for maximum benefit from the short pitch ball of bowler (Moodley, Haar & Noorbhai, 2022). The speed of body changing position and faster arms movement makes pull shot of batter more successful (Khan, Hassan, Farooq & Khan, 2018).…”

“…The good cricket pulls shot has positive link with the angular kinematics of elbows and allied joints (Noakes & Noorbhai, 2015). Furthermore, it was reported that downward pull shot in cricket is directly related with faster elbow joint extension (Moodley et al, 2022).…”

“…It is prominent from the review of literature that quantitative analysis of batters pull shot in cricket is crucial and need to be compared with kinematics of successful and unsuccessful shots. For this purpose, videography analysis is most suitable mode to better explain the kinematics of the batters pull shot technique used previously in cricket (Moodley, 2022). Researcher conducted present research on Pakistan universities cricket players representing Gomal University, Dera Ismail Khan, and (KP) Pakistan.…”