Spatio-Visual Fusion-Based Person Re-Identification for Overhead Fisheye Images
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 cl...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10121762/ |
| _version_ | 1797825674993467392 |
|---|---|
| author | Mertcan Cokbas Prakash Ishwar Janusz Konrad |
| author_facet | Mertcan Cokbas Prakash Ishwar Janusz Konrad |
| author_sort | Mertcan Cokbas |
| collection | DOAJ |
| description | 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. |
| first_indexed | 2024-03-13T10:57:18Z |
| format | Article |
| id | doaj.art-5f7204a6488446aeb4facc77fc9cd7c9 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-03-13T10:57:18Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-5f7204a6488446aeb4facc77fc9cd7c92023-05-16T23:00:20ZengIEEEIEEE Access2169-35362023-01-0111460954610610.1109/ACCESS.2023.327460010121762Spatio-Visual Fusion-Based Person Re-Identification for Overhead Fisheye ImagesMertcan Cokbas0https://orcid.org/0000-0002-6531-7653Prakash Ishwar1https://orcid.org/0000-0002-2621-1549Janusz Konrad2https://orcid.org/0000-0001-9283-5416Department of Electrical and Computer Engineering, Boston University, Boston, MA, USADepartment of Electrical and Computer Engineering, Boston University, Boston, MA, USADepartment of Electrical and Computer Engineering, Boston University, Boston, MA, USAPerson 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.https://ieeexplore.ieee.org/document/10121762/Person re-identificationfisheyeCNNdeep learningcolor histogrampeople counting |
| spellingShingle | Mertcan Cokbas Prakash Ishwar Janusz Konrad Spatio-Visual Fusion-Based Person Re-Identification for Overhead Fisheye Images IEEE Access Person re-identification fisheye CNN deep learning color histogram people counting |
| title | Spatio-Visual Fusion-Based Person Re-Identification for Overhead Fisheye Images |
| title_full | Spatio-Visual Fusion-Based Person Re-Identification for Overhead Fisheye Images |
| title_fullStr | Spatio-Visual Fusion-Based Person Re-Identification for Overhead Fisheye Images |
| title_full_unstemmed | Spatio-Visual Fusion-Based Person Re-Identification for Overhead Fisheye Images |
| title_short | Spatio-Visual Fusion-Based Person Re-Identification for Overhead Fisheye Images |
| title_sort | spatio visual fusion based person re identification for overhead fisheye images |
| topic | Person re-identification fisheye CNN deep learning color histogram people counting |
| url | https://ieeexplore.ieee.org/document/10121762/ |
| work_keys_str_mv | AT mertcancokbas spatiovisualfusionbasedpersonreidentificationforoverheadfisheyeimages AT prakashishwar spatiovisualfusionbasedpersonreidentificationforoverheadfisheyeimages AT januszkonrad spatiovisualfusionbasedpersonreidentificationforoverheadfisheyeimages |