Published Paper Details:

This paper presents a Context-Aware Framework for Real-Time Recognition of Royal Challengers Bangalore (RCB) players using hybrid deep learning techniques for intelligent sports analytics. The proposed system integrates Viola-Jones face detection, Linear Discriminant Analysis (LDA), and AdaBoost classification to accurately identify players in dynamic cricket environments. Unlike conventional face recognition methods that depend solely on facial features, the framework incorporates contextual player identification to enhance recognition performance under challenging conditions such as varying illumination, occlusion, pose changes, facial expressions, and broadcast camera variations. A comprehensive dataset comprising images of RCB players collected from match broadcasts and public sources was utilized for training and evaluation. Facial features extracted using LDA are classified through AdaBoost, while an enhanced VGG19-CNN model is employed for comparative analysis. Experimental results demonstrate that the proposed framework achieves 91% recognition accuracy, whereas the enhanced VGG19-CNN model attains 95% accuracy with superior precision, recall, and F1-score values. The effectiveness of the system is illustrated through real-time recognition of RCB players, including Virat Kohli during the IPL 2026 final, from live broadcast footage. Furthermore, a Flask-based interface enables real-time player identification, making the framework suitable for sports broadcasting, player tracking, performance analysis, and intelligent cricket analytics applications.