Person re-identification

A retrieval tool that helps a human operator browse a network of cameras is being developed. This tool addresses the problem of person re-identification: determining whether a particular individual has already appeared over a network of cameras.

Our tool allows a human operator to browse images of people extracted from a network of cameras: to detect a person on one camera and to re-detect the same person a few minutes later on another camera. The main stream is displayed on the left of the screen, while retrieval results are shown on the right. The results show lists of the most similar signatures extracted from each camera (green boxes indicate the correctly retrieved person). Below the main stream window, a topology of the camera network is shown. Detection and single camera tracking (see the main stream) are fully automatic. The human operator only selects a person of interest to produce retrieval results (right screen). The operator can easily view the original video content of retrieval results.

Retrieval in a network of cameras

credits to Sofia Zaidenberg and Bernard Boulay

Our new retrieval tool for searching a person of interest in a network of cameras. Additionally to well known re-identification feature, it allows a user to extract a trajectory of the target. This trajectory provides a full history of person appearance and it can be displayed on the map. Moreover, we provide people counter that can estimate a number of individuals w.r.t. to the given time slot (see the bar chart in the bottom screen).

Multi-camera tracking

credits to Sofia Zaidenberg and Bernard Boulay