HYDERABAD, India — Woxsen University’s Artificial Intelligence Research Centre (AIRC) and Kuwait College of Science and Technology (KCST) have jointly published a study on using artificial intelligence to predict deceptive movements in professional cricket. The research, titled ‘Using Artificial Intelligence to Predict the Next Deceptive Movement Based on Video Sequence Analysis: A Case Study on a Professional Cricket Player’s Movements’, has been featured in the Journal of Engineering Research.
The study explores how AI and machine learning can enhance sports analytics by analysing and predicting deceptive movements in cricket. Researchers employed advanced deep learning models, such as Random Forest (RF), Decision Trees (DT), and K-Nearest Neighbor (KNN), to assess and anticipate changes in player movements with an accuracy of up to 70%. This research opens new possibilities in sports biomechanics and athlete training, offering data-driven insights for players and coaches.
The team of experts include Dr K. Hemachandran, Director AIRC and Dr Korupalli V Rajesh Kumar, Chief Technology Scientist AIRC at Woxsen University, in collaboration with A. M. Mutawa Associate Professor Kuwait University and Dr M. Murugappan Professor KCST.
This research introduces a computer vision-based AI model that analyses video sequences of professional cricket players to forecast the next deceptive movement. A key application of the study focuses on analysing the bowling performance of Indian Premier League (IPL) fast bowler Umran Malik, demonstrating how AI can provide real-time insights into biomechanical variations that impact performance.
Some of the key highlights of the study include:
AI-Powered Predictive Model: A deep
learning framework that can identify
and forecast deceptive movements in sports
using pose estimation techniques.
Enhanced Training & Performance
Evaluation: The system achieved a
70.1% accuracy rate in detecting postural
changes, reinforcing AI’s role in sports
biomechanics.
Real-World Testing: The model analysed
54 video clips from IPL matches,
assessing changes in bowlers’ posture,
arm angles, and delivery mechanics.
AI-Driven Virtual Coaching: The model
lays the groundwork for virtual coaching
tools, enabling players and coaches to
refine techniques through AI-driven analysis.
This collaboration highlights the potential role of AI in improving sports training and injury prevention by offering real-time biomechanical analysis. The AI-powered tool can help to enhance training methods across multiple sports. Future research aims to integrate real-time AI applications into live matches, enhancing both player performance and strategic decision-making.
