Relation-Based Deep Attention Network with Hybrid Memory for One-Shot Person Re-Identification
One-shot person Re-identification, which owns one labeled sample among numerous unlabeled data for each identity, is proposed to tackle the problem of the shortage of labeled data. Considering the scenarios without sufficient labeled data, it is very challenging to keep abreast of the performance of...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-07-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/21/15/5113 |
| _version_ | 1827686328061394944 |
|---|---|
| author | Runxuan Si Jing Zhao Yuhua Tang Shaowu Yang |
| author_facet | Runxuan Si Jing Zhao Yuhua Tang Shaowu Yang |
| author_sort | Runxuan Si |
| collection | DOAJ |
| description | One-shot person Re-identification, which owns one labeled sample among numerous unlabeled data for each identity, is proposed to tackle the problem of the shortage of labeled data. Considering the scenarios without sufficient labeled data, it is very challenging to keep abreast of the performance of the supervised task in which sufficient labeled samples are available. In this paper, we propose a relation-based attention network with hybrid memory, which can make full use of the global information to pay attention to the identity features for model training with the relation-based attention network. Importantly, our specially designed network architecture effectively reduces the interference of environmental noise. Moreover, we propose a hybrid memory to train the one-shot data and unlabeled data in a unified framework, which notably contributes to the performance of person Re-identification. In particular, our designed one-shot feature update mode effectively alleviates the problem of overfitting, which is caused by the lack of supervised information during the training process. Compared with state-of-the-art unsupervised and one-shot algorithms for person Re-identification, our method achieves considerable improvements of 6.7%, 4.6%, and 11.5% on Market-1501, DukeMTMC-reID, and MSMT17 datasets, respectively, and becomes the new state-of-the-art method for one-shot person Re-identification. |
| first_indexed | 2024-03-10T09:09:38Z |
| format | Article |
| id | doaj.art-ea051cb3048f403aaa2248aa447444f2 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T09:09:38Z |
| publishDate | 2021-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-ea051cb3048f403aaa2248aa447444f22023-11-22T06:10:39ZengMDPI AGSensors1424-82202021-07-012115511310.3390/s21155113Relation-Based Deep Attention Network with Hybrid Memory for One-Shot Person Re-IdentificationRunxuan Si0Jing Zhao1Yuhua Tang2Shaowu Yang3State Key Laboratory of High Performance Computing, College of Computer, National University of Defense Technology, Changsha 410000, ChinaState Key Laboratory of High Performance Computing, College of Computer, National University of Defense Technology, Changsha 410000, ChinaState Key Laboratory of High Performance Computing, College of Computer, National University of Defense Technology, Changsha 410000, ChinaState Key Laboratory of High Performance Computing, College of Computer, National University of Defense Technology, Changsha 410000, ChinaOne-shot person Re-identification, which owns one labeled sample among numerous unlabeled data for each identity, is proposed to tackle the problem of the shortage of labeled data. Considering the scenarios without sufficient labeled data, it is very challenging to keep abreast of the performance of the supervised task in which sufficient labeled samples are available. In this paper, we propose a relation-based attention network with hybrid memory, which can make full use of the global information to pay attention to the identity features for model training with the relation-based attention network. Importantly, our specially designed network architecture effectively reduces the interference of environmental noise. Moreover, we propose a hybrid memory to train the one-shot data and unlabeled data in a unified framework, which notably contributes to the performance of person Re-identification. In particular, our designed one-shot feature update mode effectively alleviates the problem of overfitting, which is caused by the lack of supervised information during the training process. Compared with state-of-the-art unsupervised and one-shot algorithms for person Re-identification, our method achieves considerable improvements of 6.7%, 4.6%, and 11.5% on Market-1501, DukeMTMC-reID, and MSMT17 datasets, respectively, and becomes the new state-of-the-art method for one-shot person Re-identification.https://www.mdpi.com/1424-8220/21/15/5113hybrid memoryattentionRe-identificationone shot |
| spellingShingle | Runxuan Si Jing Zhao Yuhua Tang Shaowu Yang Relation-Based Deep Attention Network with Hybrid Memory for One-Shot Person Re-Identification Sensors hybrid memory attention Re-identification one shot |
| title | Relation-Based Deep Attention Network with Hybrid Memory for One-Shot Person Re-Identification |
| title_full | Relation-Based Deep Attention Network with Hybrid Memory for One-Shot Person Re-Identification |
| title_fullStr | Relation-Based Deep Attention Network with Hybrid Memory for One-Shot Person Re-Identification |
| title_full_unstemmed | Relation-Based Deep Attention Network with Hybrid Memory for One-Shot Person Re-Identification |
| title_short | Relation-Based Deep Attention Network with Hybrid Memory for One-Shot Person Re-Identification |
| title_sort | relation based deep attention network with hybrid memory for one shot person re identification |
| topic | hybrid memory attention Re-identification one shot |
| url | https://www.mdpi.com/1424-8220/21/15/5113 |
| work_keys_str_mv | AT runxuansi relationbaseddeepattentionnetworkwithhybridmemoryforoneshotpersonreidentification AT jingzhao relationbaseddeepattentionnetworkwithhybridmemoryforoneshotpersonreidentification AT yuhuatang relationbaseddeepattentionnetworkwithhybridmemoryforoneshotpersonreidentification AT shaowuyang relationbaseddeepattentionnetworkwithhybridmemoryforoneshotpersonreidentification |