New Trend in Video Foreground Detection Using Deep Learning

Video foreground $(FG)$ detection is one of the primitive tasks in computer vision (Cv). It is a core component in a myriad of video-based applications, like autonomous driving, activity recognition, and video surveillance. As in any other field, the deep learning (DL) has shown significant impact on video FG detection. Thus, this paper attempts to provide a fundamental study on the new strategies that harness the efficacy of deep convolutional neural networks (DCNNs) to extract video FG information. It also introduces two models and analyses their performances on various indoor and outdoor video sequences.