Crowd counting is the task of estimating the number of individuals in a crowd that has gained significant attention in computer vision research due to its diverse applications in crowd management, urban planning, and public safety. In recent years, several deep learning techniques have been developed to address several challenges like occlusion, scale variation, and illumination. This research paper aims to compare two popular crowd counting techniques namely MaskR-CNN where Resnet 50 is used as the backbone for feature extractor and Convolution Neural Network (CNN). To evaluate and compare the performance of these techniques, we use mean absolute error (MAE) and mean square error (MSE) metrics. Additionally, we analyze the computational efficiency in terms of processing time required for crowd counting on a given dataset. The current work contributes to the understanding of the trade-offs between accuracy and computational efficiency in crowd counting techniques. The findings can guide researchers and practitioners in selecting an appropriate method based on specific requirements and constraints.