Video Computing Group

Lab Group
header-image_0001_layer_4.jpg
header-image_0003_layer_0.jpg
header-image_0002_layer_1.jpg

Wide Area Scene Analysis in Vision Network

Networks of cameras are nowadays installed in various places. However, most of the data collected by such networks are analyzed manually, thus severely limiting their application. In this research, which is divided into different sub-projects, we are studying the core scientific and technical issues related to camera networks, such as information fusion, network control, activity analysis, the interplay between networking and computer vision, and computational complexity.

Camera Network Tracking and Re-identification

Camera_Network_Tracking_and_Re-identification

In many computer vision tasks it is often desirable to identify and monitor people as they move through a network of non-overlapping cameras. This is especially challenging as for a network of cameras issues such as changes of scale, illumination, viewing angle and pose start to arise. In this project we try to address this inter camera person association problem. We have shown that incorporating a consistency requirement of re-identification results across the camera network can significantly improve the re-identification performance even in the challenging scenarios where the illumination changes between the cameras are large.

Distributed Estimation

Distributed Estimation

In this project, we are developing methods for estimation and control in distributed camera networks. Specifically, we have looked at the following problems.

  • Distributed estimation algorithms that are aware of the specific constraints of camera networks
  • Distributed control for collaborative and opportunistic sensing in wide-area camera networks

Demos

Information-Weighted Consensus in a Distributed Camera Network

Active Sensing

We are developing methods for optimizing the image acquisition capabilities of the cameras so as to maximize the performance of the methods used to analyze these image. This needs to be done through collaboration between the cameras in a network, as each camera's parameters entail constraints on the others. We have developed distributed control methods for collaborative and opportunistic sensing in wide-area camera networks.

Parts of this work have been supported by the NSF CPS program under the project Distributed Sensing, Learning and Control in Dynamic Environments.