Human segmentation in surveillance video with deep learning
- PDF / 4,240,080 Bytes
- 25 Pages / 439.642 x 666.49 pts Page_size
- 106 Downloads / 290 Views
Human segmentation in surveillance video with deep learning Monica Gruosso1 · Nicola Capece2 · Ugo Erra1 Received: 29 August 2019 / Revised: 4 June 2020 / Accepted: 21 July 2020 / © The Author(s) 2020
Abstract Advanced intelligent surveillance systems are able to automatically analyze video of surveillance data without human intervention. These systems allow high accuracy of human activity recognition and then a high-level activity evaluation. To provide such features, an intelligent surveillance system requires a background subtraction scheme for human segmentation that captures a sequence of images containing moving humans from the reference background image. This paper proposes an alternative approach for human segmentation in videos through the use of a deep convolutional neural network. Two specific datasets were created to train our network, using the shapes of 35 different moving actors arranged on background images related to the area where the camera is located, allowing the network to take advantage of the entire site chosen for video surveillance. To assess the proposed approach, we compare our results with an Adobe Photoshop tool called Select Subject, the conditional generative adversarial network Pix2Pix, and the fully-convolutional model for real-time instance segmentation Yolact. The results show that the main benefit of our method is the possibility to automatically recognize and segment people in videos without constraints on camera and people movements in the scene (Video, code and datasets are available at http://graphics.unibas.it/www/HumanSegmentation/index.md.html). Keywords Deep learning · Convolutional neural network · Image processing · Background subtraction · Semantic segmentation
Ugo Erra
[email protected] Monica Gruosso [email protected] Nicola Capece [email protected] 1
Department of Mathematics, Computer Science, and Economics, University of Basilicata, Potenza, Italy
2
School of Engineering, University of Basilicata, Potenza, Italy
Multimedia Tools and Applications
1 Introduction Fully-autonomous video analysis systems have become increasingly important in recent years [1, 21, 52]. The British Security Industry Association estimated that between 4.1 and 5.9 million closed-circuit televisions were installed in the UK in 2013. Both public and private surveillance systems around the world produce an enormous amount of data, which creates a challenge for big data and artificial intelligence. There are various applications related to intelligent video surveillance, such as human search, facial recognition, people counting, and vehicles detection. Today, traditional surveillance systems are being complemented and even replaced by advanced intelligent surveillance systems. These enable high accuracy monitoring, such as human activity recognition [40]. Human activity recognition system enables continuous monitoring of human behaviors in the area of surveillance, which allow tracking of human body parts such as head, torso, arms, and legs to perform activity recognit
Data Loading...