Saliency Based Person Re-Identification in Video Using Color Features

Human re-identification is to match persons observed in non-overlapping camera views with visual features for inter camera tracking. Person re-identification in a non-overlapping multi-camera scenario is an open and interesting challenge. Humans are inherently able to sample those relevant people's details that allow us to correctly solve the problem in a fraction of a second. But the task can hardly be completed by machines. Human salience is distinctive and reliable information in matching persons across disjoint camera views. So a saliency based approach can be used to re-identify persons in video. The model includes three stages: I) input processing, ii) video processing, iii) matching. The proposed person representation combines visual features either considering or not the saliency. The proposed approach has been extensively evaluated on 3DPes public datasets.