Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Vision-based Vehicle Re-identification in Bridge Scenario using Flock Similarity

Chunfeng Zhang, Ping Wang

TL;DR

A vehicle re-identification method based on flock similarity is presented, which improves the accuracy of vehicle re-identification by utilizing vehicle information adjacent to the target vehicle when the relative position of the vehicles remains unchanged and flock size is appropriate.

Abstract

Due to the needs of road traffic flow monitoring and public safety management, video surveillance cameras are widely distributed in urban roads. However, the information captured directly by each camera is siloed, making it difficult to use it effectively. Vehicle re-identification refers to finding a vehicle that appears under one camera in another camera, which can correlate the information captured by multiple cameras. While license plate recognition plays an important role in some applications, there are some scenarios where re-identification method based on vehicle appearance are more suitable. The main challenge is that the data of vehicle appearance has the characteristics of high inter-class similarity and large intra-class differences. Therefore, it is difficult to accurately distinguish between different vehicles by relying only on vehicle appearance information. At this time, it is often necessary to introduce some extra information, such as spatio-temporal information. Nevertheless, the relative position of the vehicles rarely changes when passing through two adjacent cameras in the bridge scenario. In this paper, we present a vehicle re-identification method based on flock similarity, which improves the accuracy of vehicle re-identification by utilizing vehicle information adjacent to the target vehicle. When the relative position of the vehicles remains unchanged and flock size is appropriate, we obtain an average relative improvement of 204% on VeRi dataset in our experiments. Then, the effect of the magnitude of the relative position change of the vehicles as they pass through two cameras is discussed. We present two metrics that can be used to quantify the difference and establish a connection between them. Although this assumption is based on the bridge scenario, it is often true in other scenarios due to driving safety and camera location.

Vision-based Vehicle Re-identification in Bridge Scenario using Flock Similarity

TL;DR

A vehicle re-identification method based on flock similarity is presented, which improves the accuracy of vehicle re-identification by utilizing vehicle information adjacent to the target vehicle when the relative position of the vehicles remains unchanged and flock size is appropriate.

Abstract

Due to the needs of road traffic flow monitoring and public safety management, video surveillance cameras are widely distributed in urban roads. However, the information captured directly by each camera is siloed, making it difficult to use it effectively. Vehicle re-identification refers to finding a vehicle that appears under one camera in another camera, which can correlate the information captured by multiple cameras. While license plate recognition plays an important role in some applications, there are some scenarios where re-identification method based on vehicle appearance are more suitable. The main challenge is that the data of vehicle appearance has the characteristics of high inter-class similarity and large intra-class differences. Therefore, it is difficult to accurately distinguish between different vehicles by relying only on vehicle appearance information. At this time, it is often necessary to introduce some extra information, such as spatio-temporal information. Nevertheless, the relative position of the vehicles rarely changes when passing through two adjacent cameras in the bridge scenario. In this paper, we present a vehicle re-identification method based on flock similarity, which improves the accuracy of vehicle re-identification by utilizing vehicle information adjacent to the target vehicle. When the relative position of the vehicles remains unchanged and flock size is appropriate, we obtain an average relative improvement of 204% on VeRi dataset in our experiments. Then, the effect of the magnitude of the relative position change of the vehicles as they pass through two cameras is discussed. We present two metrics that can be used to quantify the difference and establish a connection between them. Although this assumption is based on the bridge scenario, it is often true in other scenarios due to driving safety and camera location.
Paper Structure (5 sections, 10 equations, 9 figures, 1 table)

This paper contains 5 sections, 10 equations, 9 figures, 1 table.

Table of Contents

  1. Introduction
  2. Flock Similarity
  3. Prepare for experiments
  4. Experiments
  5. Conclusion

Figures (9)

  • Figure 1: Some examples from VeRi dataset. Left: different vehicles with a highly similar appearance. Right: the same vehicle from different viewpoints.
  • Figure 2: Scene of some vehicles passing through two cameras in the bridge scenario. The relative position between them rarely changes.
  • Figure 3: A simple schematic representation of vehicles passing through two cameras.
  • Figure 4: A simple example of flock similarity.
  • Figure 5: The Siamese Neural Network architecture.
  • ...and 4 more figures