Batch Hard Contrastive Loss and Its Application to Cross-View Gait Recognition
Biometric person authentication comprises two tasks: the identification task (i.e., one-to-many matching) and the verification task (i.e., one-to-one matching). In this paper, we propose a loss function called batch hard contrastive loss (BHCn) for the deep learning-based verification task. For this...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10082933/ |
| _version_ | 1827976122434846720 |
|---|---|
| author | Mohamad Ammar Alsherfawi Aljazaerly Yasushi Makihara Daigo Muramatsu Yasushi Yagi |
| author_facet | Mohamad Ammar Alsherfawi Aljazaerly Yasushi Makihara Daigo Muramatsu Yasushi Yagi |
| author_sort | Mohamad Ammar Alsherfawi Aljazaerly |
| collection | DOAJ |
| description | Biometric person authentication comprises two tasks: the identification task (i.e., one-to-many matching) and the verification task (i.e., one-to-one matching). In this paper, we propose a loss function called batch hard contrastive loss (BHCn) for the deep learning-based verification task. For this purpose, we consider batch mining techniques developed in the identification task and translate them to the verification task. More specifically, inspired by batch mining triplet losses to learn a relative distance for the identification task, we propose BHCn to learn an absolute distance that better represents verification in general. Our method preserves the identity-agnostic nature of the contrastive loss by selecting the hardest pair of samples for each pair of identities in a batch instead of selecting the hardest pair for each sample. We validate the effectiveness of the proposed method in cross-view gait recognition using three networks: a lightweight input, structure, and output network we call GEI + CNN (Gait Energy Image Convolutional Neural Network) as well as the widely used GaitSet and GaitGL, which have sophisticated inputs, structures, and outputs. We trained these networks with the publicly available silhouette-based datasets, the OU-ISIR Gait Database Multi-View Large Population (OU-MVLP) dataset and the Institute of Automation Chinese Academy of Sciences Gait Database Multiview (CASIA-B) dataset. Experimental results show that the proposed BHCn outperforms other loss functions, such as a triplet loss with batch mining as well as the conventional contrastive loss. |
| first_indexed | 2024-04-09T20:22:28Z |
| format | Article |
| id | doaj.art-bf1fc61c2d2c424b8f5f8634a5541766 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-09T20:22:28Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-bf1fc61c2d2c424b8f5f8634a55417662023-03-30T23:01:51ZengIEEEIEEE Access2169-35362023-01-0111311773118710.1109/ACCESS.2023.326227110082933Batch Hard Contrastive Loss and Its Application to Cross-View Gait RecognitionMohamad Ammar Alsherfawi Aljazaerly0https://orcid.org/0000-0003-1912-804XYasushi Makihara1Daigo Muramatsu2https://orcid.org/0000-0002-0053-4868Yasushi Yagi3https://orcid.org/0000-0002-3546-8071Department of Intelligent Media, Institute of Scientific and Industrial Research, Osaka University, Osaka, JapanDepartment of Intelligent Media, Institute of Scientific and Industrial Research, Osaka University, Osaka, JapanFaculty of Science and Technology, Seikei University, Musashino, Tokyo, JapanDepartment of Intelligent Media, Institute of Scientific and Industrial Research, Osaka University, Osaka, JapanBiometric person authentication comprises two tasks: the identification task (i.e., one-to-many matching) and the verification task (i.e., one-to-one matching). In this paper, we propose a loss function called batch hard contrastive loss (BHCn) for the deep learning-based verification task. For this purpose, we consider batch mining techniques developed in the identification task and translate them to the verification task. More specifically, inspired by batch mining triplet losses to learn a relative distance for the identification task, we propose BHCn to learn an absolute distance that better represents verification in general. Our method preserves the identity-agnostic nature of the contrastive loss by selecting the hardest pair of samples for each pair of identities in a batch instead of selecting the hardest pair for each sample. We validate the effectiveness of the proposed method in cross-view gait recognition using three networks: a lightweight input, structure, and output network we call GEI + CNN (Gait Energy Image Convolutional Neural Network) as well as the widely used GaitSet and GaitGL, which have sophisticated inputs, structures, and outputs. We trained these networks with the publicly available silhouette-based datasets, the OU-ISIR Gait Database Multi-View Large Population (OU-MVLP) dataset and the Institute of Automation Chinese Academy of Sciences Gait Database Multiview (CASIA-B) dataset. Experimental results show that the proposed BHCn outperforms other loss functions, such as a triplet loss with batch mining as well as the conventional contrastive loss.https://ieeexplore.ieee.org/document/10082933/Biometricsdeep learningforensicsgait recognition |
| spellingShingle | Mohamad Ammar Alsherfawi Aljazaerly Yasushi Makihara Daigo Muramatsu Yasushi Yagi Batch Hard Contrastive Loss and Its Application to Cross-View Gait Recognition IEEE Access Biometrics deep learning forensics gait recognition |
| title | Batch Hard Contrastive Loss and Its Application to Cross-View Gait Recognition |
| title_full | Batch Hard Contrastive Loss and Its Application to Cross-View Gait Recognition |
| title_fullStr | Batch Hard Contrastive Loss and Its Application to Cross-View Gait Recognition |
| title_full_unstemmed | Batch Hard Contrastive Loss and Its Application to Cross-View Gait Recognition |
| title_short | Batch Hard Contrastive Loss and Its Application to Cross-View Gait Recognition |
| title_sort | batch hard contrastive loss and its application to cross view gait recognition |
| topic | Biometrics deep learning forensics gait recognition |
| url | https://ieeexplore.ieee.org/document/10082933/ |
| work_keys_str_mv | AT mohamadammaralsherfawialjazaerly batchhardcontrastivelossanditsapplicationtocrossviewgaitrecognition AT yasushimakihara batchhardcontrastivelossanditsapplicationtocrossviewgaitrecognition AT daigomuramatsu batchhardcontrastivelossanditsapplicationtocrossviewgaitrecognition AT yasushiyagi batchhardcontrastivelossanditsapplicationtocrossviewgaitrecognition |