The Role of Traditional Methods in Deep Learning Background Modelling

We express our gratitude for your insightful comments on our paper and the valuable suggestions you provided concerning the utilization of deep learning frameworks for background modelling. Although we acknowledge the potential of deep learning algorithms in various image processing tasks, we have observed that traditional methods do not necessarily conflict with deep learning frameworks. For instance, Tezcan M O, et al. proposed the BSUV-Net Algorithm for unseen videos based on a full convolutional neural network, as documented in the literature. In their approach, they employ two reference frames to characterize the background. One frame is an Empty 'background frame', with no people or other objects of interest, while the other reference frame characterizes recent background by computing the media of 100 frames preceding the frame being processed. This implies that the background generated by traditional methods can also be incorporated as a part of deep learning frameworks. Therefore, it is meaningful to construct the background based on traditional methods.