Special issue on “DLVA: Advances in Deep Learning and Visual Analytics for Intelligent Surveillance Systems”
摘要截稿:
全文截稿: 2018-09-30
影响因子: 3.255
期刊难度:
CCF分类: C类
中科院JCR分区:
• 大类 : 计算机科学 - 3区
• 小类 : 计算机:人工智能 - 3区
Overview
The increasing sophistication and diversity of threats to public security have been calling critical demand of developing and deploying reliable, secure, and timely efficient visual intelligent surveillance systems in smart cities. For example, visual surveillance for indoor environments, like metro stations, plays an important role both in the assurance of safety conditions for the public and in the management of the transport network. When designing the next generation security solutions, it is crucial to combine sensing, computing, understanding, communication and prediction in such networked-camera systems. Examples include automated video surveillance platforms and smart camera networked systems that are monitoring the behavior, activities, or other changing information for the purpose of influencing, managing, directing, or protecting people. They exhibit a high-level of awareness beyond primitive actions, in support of persistent and long-term autonomy. However, some core problems such as object identification and tracking, and behavior analysis in intelligent surveillance are still affected by a number of practical problems. They typically involve a variety of representation, reasoning and efficiency mechanisms in the context of an extended distance and period of time and low resolution/frame rate in poor quality capturing conditions. Recent progress in computer vision techniques and related visual analytics offers new prospects for an intelligent surveillance system. A major recent development is the massive success resulting from using the deep learning techniques to enable the significant boosting of visual analysis performance and initiate new research directions to understand visual content. For example, convolutional neural networks have demonstrated superiority on modeling high-level visual concepts, while recurrent neural networks have shown promise in modeling temporal dynamics in videos. It has been and will be seen as resolution to change the whole visual recognition systems. It is expected that the development of deep learning and its related visual analytic methodologies would further influence the field of intelligent surveillance systems.
This special issue will serve a platform to publish state-of-the-art advancements in this domain of research and seeks for original contributions of work, which addresses the challenges from using deep learning and related techniques to understand and promote the ubiquitous intelligent surveillance systems. Original papers to survey the recent progress in this exciting area and highlight potential solutions to common challenging problems are also welcome. The list of possible topics includes, but not limited to:
- Emotion/Gait/Activity/Gesture recognition and prediction
- Large-scale video indexing
- Pedestrian detection in the wild
- Scene understanding and human behavior analysis
- Person re-identification and biometric recognition
- Summarization of long surveillance videos
- Visual analytics for forensics and security applications
- Pedestrian and vehicle navigation tracking
- Face recognition and verification
- Event (abnormal) detection and recognition
- Cloud and distributed for visual surveillance
- Object tracking and segmentation
- Human computer/robot interactions
- Data collections, benchmarking and performance evaluations