Joint Person Re-identification and Camera Network Topology Inference
in Multiple Cameras
Yeong-Jun Cho¹², Su-A Kim¹, Jae-Han Park¹, Kyuewang Lee¹ and Kuk-Jin Yoon²
¹Gwangju Institute of Science and Technology (GIST),
²Korea Advanced Institute of Science and Technology (KAIST)
Computer Vision and Image Understanding (CVIU) 2019
IEEE/CVF International Conference on Computer Vision Workshops (ICCVW) 2017
[Paper-journal] [Paper-conference] [Workshop page] [Code] [Dataset]
Notice
The codes and datasets are very old version (written in MATLAB). No maintenance for them. Sorry for this...!😿
Please refer to our recent work "Re-identification Based on Spatial-temporal Fusion Network" [project]
ABSTRACT
In this study, we propose a unified framework which jointly solves both person re-identification and camera network topology inference problems with minimal prior knowledge about the environments. The proposed framework takes general multi-camera network environments into account and can be applied to online person re-identification in large-scale multi-camera networks. In addition, to show the superiority of the proposed framework, we provide a new person re-identification dataset with full annotations, named SLP, captured in the multi-camera network. Experimental results using our re-identification and public datasets show that the proposed methods are promising for both person re-identification and camera topology inference tasks.
OVERVIEW
The main idea of this study is that the camera network topology inference and person re-identification can be achieved jointly while complementing each other. Based on this idea, we propose a unified framework, which automatically solves both the problems together. To the best of our knowledge, this is the first attempt to solve both the problems jointly. In the proposed framework, we first infer the initial camera network topology using only highly reliable re-identification results obtained by the proposed multi-shot matching method. This initial topology is used to improve the person re-identification results, and the improved re-identification results are then used to refine the camera network topology. This procedure is repeated until the estimated camera network topology converges. Once we estimate the reliable camera network topology in the initialization stage, we utilize it for the online person re-identification and update the camera network topology with time.
Figure 1. The proposed unified framework for person re-identification and camera network topology inference
RESULTS
We qualitatively and quantitatively evaluated and compared the performance of the proposed framework with state-of-the-art methods using public and our own datasets. The results show that the proposed framework is promising for both person re-identification and camera topology inference and superior to other frameworks in terms of both speed and accuracy.
Figure 2. The iterative update of connectivity confidences. The connectivity of each link has strengthened during the iterations
Table 1. Performance comparison with other methods in offline initialization stage
ABSTRACT
This is an improved research from the work "Joint Person Re-identification and Camera Network Topology Inference"
In this study, we propose a novel distance-based camera network topology inference method for efficient person re-identification. We first calibrate each camera, and estimate the relative scales between cameras. Using the calibration results for multiple cameras, we estimate the speed of each person, and infer the distances between cameras to generate a distance-based camera network topology. The proposed distance-based topology can be adaptively applied to each person according to their speed, and can handle diverse transition times of people between non-overlapping cameras. To validate the proposed method, we test it using open person re-identification datasets, and compare the results with state-of-the-art methods. The experimental results demonstrate that the proposed method is effective for person re-identification in a large-scale camera network with various transition times for people.
OVERVIEW
When a person walks significantly faster or slower than the average speed, it is difficult to precisely predict their transition time between two non-overlapping cameras based on a time-based topology, and this makes it difficult to restrict or reduce the solution space. To overcome the limitations of previous studies, we compute the speeds of people and exploit this speed information to in fer a new distance-based camera network topology.
Figure 3. Overview of the proposed framework: distance-based camera network topology inference and person re-identification
RESULTS
Based on the distance-based camera network topology, we enhanced both accuracy of topology and re-identification
Figure 4. Topology and re-identfication accuracies according to average search range
🖇️ Citation
@article{cho2019joint,
title={Joint person re-identification and camera network topology inference in multiple cameras},
author={Cho, Yeong-Jun and Kim, Su-A and Park, Jae-Han and Lee, Kyuewang and Yoon, Kuk-Jin},
journal={Computer Vision and Image Understanding}, volume={180}, pages={34--46},
year={2019}, publisher={Elsevier} }
@article{cho2019distance,
title={Distance-based camera network topology inference for person re-identification},
author={Cho, Yeong-Jun and Yoon, Kuk-Jin},
journal={Pattern Recognition Letters}, volume={125}, pages={220--227},
year={2019}, publisher={Elsevier} }
