Cricket Batting Analysis Using Pose Estimations and Deep Neural Network

Abstract

When it comes to analysis, it’s always being a complicated and challenging process. AI based Batting technique analysis has been an important area due to evaluation of the sport with the technology. This report is based on the research conducted in Cricket batting stroke analysis and suggesting an alternative system did by using CNN (convolutional neural networks) and pose estimation models. The models that were trained using CNN based architectures like resnet50 and residual network given the accuracy over 75%. All the relevant code snippets, charts and tables are illustrated in the relevant chapters. The proposed system has been evaluated and tested and all the test results, design, implementation and documentation are expressed in an efficient manner