Spatio-Visual Fusion-Based Person Re-Identification for Overhead Fisheye Images

TL;DR

This paper introduces a multi-feature fusion framework for person re-identification using overhead fisheye cameras, significantly improving accuracy over existing appearance-based methods and demonstrating potential for crowd counting applications.

Contribution

It proposes a novel multi-feature fusion approach combining deep-learning, color, and location features specifically for overhead fisheye PRID, addressing a performance gap in current methods.

Findings

Outperforms recent appearance-based deep-learning methods by 18% in accuracy

Outperforms location-based methods by 3% in accuracy

Shows potential for crowd counting in indoor spaces

Abstract

Person re-identification (PRID) has been thoroughly researched in typical surveillance scenarios where various scenes are monitored by side-mounted, rectilinear-lens cameras. To date, few methods have been proposed for fisheye cameras mounted overhead and their performance is lacking. In order to close this performance gap, we propose a multi-feature framework for fisheye PRID where we combine deep-learning, color-based and location-based features by means of novel feature fusion. We evaluate the performance of our framework for various feature combinations on FRIDA, a public fisheye PRID dataset. The results demonstrate that our multi-feature approach outperforms recent appearance-based deep-learning methods by almost 18% points and location-based methods by almost 3% points in matching accuracy. We also demonstrate the potential application of the proposed PRID framework to people counting in large, crowded indoor spaces.