Hand Gesture Recognition: Enhancing Accuracy and Precision with Deep Learning

Accurate, real-time recognition of hand gestures in dynamic environments remains challenging in human-computer interaction. This paper presents a hybrid deep learning model combining Convolutional Neural Networks (CNN) with Recurrent Neural Networks (RNN) using Long Short-Term Memory (LSTM) layers to capture both spatial and temporal information for dynamic hand gesture recognition. Trained on a dataset of six gestures-scroll-left, scroll-right, scroll-up, scroll-down, zoom-in, and zoom-out-the model achieves an accuracy of 94.66%, with an F1 score of 0.94 and an AUC-ROC of 0.95, indicating significant improvement over traditional models and practical viability in real-world applications. Key topics include data preprocessing, model architecture, hardware and software configurations, and performance comparisons with benchmarks. The paper concludes with discussions on limitations and future research directions to enhance the model's adaptability and efficiency.