Yoga has surged in popularity as a comprehensive method for enhancing physical, mental, and spiritual well-being. With this growing interest, there is an increasing demand for effective techniques to supervise and monitor personal yoga sessions. In this study, we introduce a Yoga Pose Detection (YPD) method that harnesses VGG16, a Convolutional Neural Network (CNN) model, to accurately recognize yoga poses. Prevailing methods for yoga pose recognition often face challenges in complex real-world contexts due to their reliance on manually engineered features derived from controlled environments. The proposed YPD model effectively overcomes this challenge by fine-tuning the pre-trained VGG16 model using a new dataset specially curated for yoga poses. The efficacy of the proposed approach is validated on the publicly available Yoga-82 dataset, attaining an impressive accuracy of 97.83%. Additionally, the proposed model surpasses numerous state-of-the-art image classification models, including ResNet 50, Inception V3, Inception ResNet-V2, Xception, Classic CNN, and Transfer Learning & Stochastic Gradient Descent (SGD) in terms of accuracy, precision, recall, and F1 score.