Integrated Single Shot Multi-Box Detector and Efficient Pre-Trained Deep Convolutional Neural Network for Partially Occluded Face Recognition System
Partial occlusion is a key issue in face recognition as it decreases the recognition accuracy. As a result, current face recognition systems are limited to operate under constrained environments. To resolve the partial occlusion problem, we propose a system that adopts the convolutional neural netwo...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9638690/ |
| _version_ | 1818979538809389056 |
|---|---|
| author | An-Chao Tsai Yang-Yen Ou Wei-Ching Wu Jhing-Fa Wang |
| author_facet | An-Chao Tsai Yang-Yen Ou Wei-Ching Wu Jhing-Fa Wang |
| author_sort | An-Chao Tsai |
| collection | DOAJ |
| description | Partial occlusion is a key issue in face recognition as it decreases the recognition accuracy. As a result, current face recognition systems are limited to operate under constrained environments. To resolve the partial occlusion problem, we propose a system that adopts the convolutional neural network but with a pre-trained model for robust face recognition and facial feature extraction. The model improves the accuracy of partially occluded face recognition. Moreover, the face detection network utilizes the feature pyramids to reduce the number of network parameters and achieve scale invariance. The image context module is also incorporated to increase the receptive field, as it effectively improves the detection accuracy and reduces memory usage. Experiment results show that the proposed method performs better than existing state-of-the-art methods for the detection and recognition of occluded faces. |
| first_indexed | 2024-12-20T17:01:08Z |
| format | Article |
| id | doaj.art-25f6f15797174da7badd191f443a4457 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-20T17:01:08Z |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-25f6f15797174da7badd191f443a44572022-12-21T19:32:31ZengIEEEIEEE Access2169-35362021-01-01916414816415810.1109/ACCESS.2021.31334469638690Integrated Single Shot Multi-Box Detector and Efficient Pre-Trained Deep Convolutional Neural Network for Partially Occluded Face Recognition SystemAn-Chao Tsai0https://orcid.org/0000-0003-4364-6760Yang-Yen Ou1Wei-Ching Wu2Jhing-Fa Wang3International Master Program of Information Technology and Application, National Pingtung University, Pingtung, TaiwanDepartment of Electrical Engineering, National Cheng Kung University, Tainan, TaiwanDepartment of Electrical Engineering, National Cheng Kung University, Tainan, TaiwanDepartment of Electrical Engineering, National Cheng Kung University, Tainan, TaiwanPartial occlusion is a key issue in face recognition as it decreases the recognition accuracy. As a result, current face recognition systems are limited to operate under constrained environments. To resolve the partial occlusion problem, we propose a system that adopts the convolutional neural network but with a pre-trained model for robust face recognition and facial feature extraction. The model improves the accuracy of partially occluded face recognition. Moreover, the face detection network utilizes the feature pyramids to reduce the number of network parameters and achieve scale invariance. The image context module is also incorporated to increase the receptive field, as it effectively improves the detection accuracy and reduces memory usage. Experiment results show that the proposed method performs better than existing state-of-the-art methods for the detection and recognition of occluded faces.https://ieeexplore.ieee.org/document/9638690/Convolutional neural networkface databasepartial face recognition |
| spellingShingle | An-Chao Tsai Yang-Yen Ou Wei-Ching Wu Jhing-Fa Wang Integrated Single Shot Multi-Box Detector and Efficient Pre-Trained Deep Convolutional Neural Network for Partially Occluded Face Recognition System IEEE Access Convolutional neural network face database partial face recognition |
| title | Integrated Single Shot Multi-Box Detector and Efficient Pre-Trained Deep Convolutional Neural Network for Partially Occluded Face Recognition System |
| title_full | Integrated Single Shot Multi-Box Detector and Efficient Pre-Trained Deep Convolutional Neural Network for Partially Occluded Face Recognition System |
| title_fullStr | Integrated Single Shot Multi-Box Detector and Efficient Pre-Trained Deep Convolutional Neural Network for Partially Occluded Face Recognition System |
| title_full_unstemmed | Integrated Single Shot Multi-Box Detector and Efficient Pre-Trained Deep Convolutional Neural Network for Partially Occluded Face Recognition System |
| title_short | Integrated Single Shot Multi-Box Detector and Efficient Pre-Trained Deep Convolutional Neural Network for Partially Occluded Face Recognition System |
| title_sort | integrated single shot multi box detector and efficient pre trained deep convolutional neural network for partially occluded face recognition system |
| topic | Convolutional neural network face database partial face recognition |
| url | https://ieeexplore.ieee.org/document/9638690/ |
| work_keys_str_mv | AT anchaotsai integratedsingleshotmultiboxdetectorandefficientpretraineddeepconvolutionalneuralnetworkforpartiallyoccludedfacerecognitionsystem AT yangyenou integratedsingleshotmultiboxdetectorandefficientpretraineddeepconvolutionalneuralnetworkforpartiallyoccludedfacerecognitionsystem AT weichingwu integratedsingleshotmultiboxdetectorandefficientpretraineddeepconvolutionalneuralnetworkforpartiallyoccludedfacerecognitionsystem AT jhingfawang integratedsingleshotmultiboxdetectorandefficientpretraineddeepconvolutionalneuralnetworkforpartiallyoccludedfacerecognitionsystem |