Deep Learning in Cricket: A Comprehensive Survey of Shot Detection and Performance Analysis

Abstract

Cricket shot detection and performance analysis have become increasingly significant in the realm of sports analytics. This paper presents a comprehensive survey of recent advancements in cricket shot detection and performance analysis using deep learning techniques. Through a systematic review of ten research papers, various methodologies, algorithms, and approaches employed in the field were examined. The surveyed papers showcase the application of deep convolutional neural networks (CNNs), transfer learning, and multimodal feature integration for accurate shot detection and outcome classification. Additionally, the survey highlights common processes such as data preprocessing, model training, and evaluation metrics, while also discussing the integration of features extracted from video, audio, and image modalities. Through a detailed literature review, this paper analyzes the similarities and differences in approaches, methodologies, and results across the surveyed papers. Furthermore, this survey identifies opportunities for future research and development to advance the field of cricket shot detection and performance analysis. Overall, this survey provides valuable insights into the state-of-the-art techniques and challenges in utilizing deep learning for cricket analytics, offering guidance for researchers and practitioners in the domain.