Cooperative tracking and visual analytics
@ Dep. Telecommunications and Information Processing (Ghent University), Belgium
One of the most common but still challenging requirements in multi-camera video processing is the ability to automatically track objects over multiple cameras. In intelligent traffic management, for example, objects of interest include not only vehicles but also weak road users. Current state of the art approaches focus either on feature-modelling that designs descriptors invariant to camera changes or on metric learning that often require prohibitive amount of training data. Vehicle tracking/re-identification is equally challenging in difficult circumstances. The first goal of this project is to design algorithms for distributed multiple targets tracking through a decentralized approach. The second goal is to improve object detection and tracking using a multi-sensor approach. Thermal cameras have promising potential in surveillance applications, especially when combined with optical cameras. The third goal of the project is to provide solutions for behaviour analysis and action recognition. The research will use high-level analysis to automatically determine which cameras observe the same or similar action, such as pedestrians waiting to cross the street. Deep learning is a promising approach.
Chengjin Lyu was born in Qingdao, China in 1993. He obtained his Bachelor and Master of Engineering degrees from Department of Measurement Control and Information Technology, Beihang University, in 2015 and 2018, respectively. His main research interests focus on computer vision related topics for cooperative tracking applications. Chengjin is a researcher working on his doctorate under the supervision of Prof. Wilfried Philips, at the Image Processing and Interpretation (IPI) research group, TELIN Department at Ghent University in Belgium. IPI group is affiliated with imec.
"Cooperative tracking and visual analytics"
One of the most common but highly challenging tasks in multi-camera video processing is to automatically track objects over multiple cameras, especially in the complex conditions of road traffic. In intelligent traffic management, for example, objects of interest include not only vehicles but also vulnerable road users. An ideal smart camera system can track objects cooperatively and analyze visual information locally. With the help of cooperative tracking and visual analytics of multiple targets, we can achieve smart surveillance, which could help optimize the traffic flow, track suspects, prevent theft, etc
- Design algorithms for distributed multiple targets re-identification/tracking
- Improve object detection and tracking using multi-sensor approach
- Design methods for behavior analysis and action recognition in road traffic environment
Cooperative multi-camera algorithms, specifically in the field of cooperative object/person re-identification, tracking and visual analytics, in road traffic environments.